advancing ecological understandings through technological ...canis rufus) hybridization monitoring...
TRANSCRIPT
TECHNICAL REVIEW
Advancing ecological understandings throughtechnological transformations in noninvasive genetics
ALBANO BEJA-PEREIRA,*1 RITA OLIVEIRA,*†1 PAULO C. ALVES,*† MICHAEL K. SCHWARTZ‡
and GORDON LUIKART*§
*CIBIO, Centro de Investigacao em Biodiversidade e Recursos Geneticos, Universidade do Porto, Campus Agrario de Vairao,
4485-661 Vairao, Portugal, †Departamento de Zoologia e Antropologia, Faculdade de Ciencias da Universidade do Porto, Rua
Campo Alegre s ⁄ n, 4169-007 Porto, Portugal, ‡USDA Forest Service, Rocky Mountain Research Station, Missoula, MT 59801,
USA, §Division of Biological Sciences, University of Montana, Missoula, MT 59812, USA
Abstract
Noninvasive genetic approaches continue to improve studies in molecular ecology, conserva-
tion genetics and related disciplines such as forensics and epidemiology. Noninvasive sam-
pling allows genetic studies without disturbing or even seeing the target individuals.
Although noninvasive genetic sampling has been used for wildlife studies since the 1990s,
technological advances continue to make noninvasive approaches among the most used and
rapidly advancing areas in genetics. Here, we review recent advances in noninvasive genetics
and how they allow us to address important research and management questions thanks to
improved techniques for DNA extraction, preservation, amplification and data analysis. We
show that many advances come from the fields of forensics, human health and domestic
animal health science, and suggest that molecular ecologists explore literature from these
fields. Finally, we discuss how the combination of advances in each step of a noninvasive
genetics study, along with fruitful areas for future research, will continually increase the
power and role of noninvasive genetics in molecular ecology and conservation genetics.
Keywords: conservation genetics, forensics, genomics, molecular ecology, population genetics
Received 8 May 2008; revision accepted 17 February 2009
Introduction
Noninvasive genetic sampling was first used in wild ani-
mals nearly two decades ago (Hoss et al. 1992; Taberlet &
Bouvet 1992). The main advantage of noninvasive genet-
ics is that it allows biologists to study many individuals
and populations without contacting, disturbing, or even
seeing the organisms. Samples collected noninvasively
include faeces, hairs, urine, saliva from chewed material,
feathers, scent marks, eggshells, sloughed skin, and even
menstrual fluid (Table 1). The largest contributions of
noninvasive approaches are to studies that focus on (i)
identification of individuals for studies of population size
and individual movement, (ii) wildlife forensic cases, (iii)
delineation of populations and population genetic
parameters (structure, gene flow and demographic his-
tory such as bottleneck detection), and (iv) assessment of
mating systems and behavioural ecology (Table 1).
A growing number of noninvasive techniques yield
good enough DNA and low enough genotyping error
rates to allow researchers to address nearly all questions
that can be addressed using traditional high-quality sam-
ples such as blood (e.g. Epps et al. 2006; Luikart et al.
2008a). This is exciting because noninvasive studies
5–10 years ago were generally more limited in scope by
high genotyping error rates and low polymerase chain
reaction (PCR) amplification success (reviewed in Taber-
let et al. 1999; Waits & Paetkau 2005). In this review, we
report recent advances from different research fields,
hoping to open communication channels and diffuse
information among disciplines.
Rapid advancements in forensic science, human
medical research, and livestock disease studies, and
ancient DNA techniques continuously generate
Correspondence: Gordon Luikart, Fax: +351252661780;
E-mail: [email protected] authors contributed equally to this work.
� 2009 Blackwell Publishing Ltd
Molecular Ecology Resources (2009) 9, 1279–1301 doi: 10.1111/j.1755-0998.2009.02699.x
Tab
le1
Ali
sto
fd
iffe
ren
tb
iolo
gic
alsa
mp
les,
tax
aan
dp
urp
ose
sfo
rw
hic
hn
on
inv
asiv
esa
mp
lin
gh
ave
bee
nu
sed
inw
ild
anim
alp
op
ula
tio
ns
Gro
up
Sp
ecie
sP
urp
ose
Stu
dy
Blo
od
insn
ow
Mam
mal
sW
olf
(Can
islu
pus)
Sp
ecie
sid
enti
fica
tio
nS
can
du
ra(2
005)
Bu
ccal
and
clo
acal
swab
Am
ph
ibia
ns
and
rep
tile
sT
uat
ara
(Sph
enod
onpu
nct
atu
s)M
eth
od
olo
gy
Mil
ler
(200
6)
Bu
ccal
swab
Am
ph
ibia
ns
and
rep
tile
s
Alp
ine
new
t(T
ritu
rus
alpe
stri
s)an
dg
reen
tree
fro
g
(Hyl
aar
bore
a)
Met
ho
do
log
yB
roq
uet
etal
.(20
07)a
Bu
ccal
swab
Bir
ds
Bla
ck-c
app
edch
ick
adee
(Poe
cile
atri
capi
llu
s)an
d
bo
real
chic
kad
ee(P
oeci
lehu
dson
ica)
Met
ho
do
log
y,i
nd
ivid
ual
and
gen
der
iden
tifi
cati
on
Han
del
etal
.(20
06)
Eg
gsh
ell
Bir
ds
Gre
ater
sno
wg
oo
se(C
hen
caer
ule
scen
sat
lan
tica
)M
eth
od
olo
gy
Lec
om
teet
al.(
2006
)
Eg
gsh
ell
Bir
ds
Do
mes
tic
chic
ken
(Gal
lus
gall
us)
Ch
ick
enan
aem
iav
iru
sd
etec
tio
nM
ille
ret
al.(
2003
)
Eg
gsh
ell,
feat
her
s,
bu
ccal
swab
Bir
ds
Sag
e-G
rou
se(C
entr
ocer
cus
uro
piha
sian
us)
Gen
der
det
erm
inat
ion
Bu
shet
al.(
2005
)
Eg
gsh
ell
Bir
ds
Cas
pia
nte
rn(S
tern
aca
spia
)an
dh
erri
ng
gu
ll(L
aru
s
arge
nta
tus)
Met
ho
do
log
yS
chm
altz
etal
.(20
06)
Fae
ces
Bir
ds
Ch
ick
-rea
rin
gm
acar
on
ip
eng
uin
(Eu
dypt
es
chry
solo
phu
s)
Die
td
eter
min
atio
nD
eag
leet
al.(
2007
)
Fae
ces
Bir
ds
Eu
rop
ean
sto
nec
hat
(Sax
icol
ato
rqu
ata
rubi
cola
)H
orm
on
esm
on
ito
rin
gG
oy
man
n(2
005)
Fae
ces
Bir
ds
Cap
erca
illi
e(T
etra
ou
roga
llu
s)P
op
ula
tio
nst
ruct
ure
,gen
efl
ow
Reg
nau
tet
al.(
2006
)
Fae
ces
Mam
mal
sR
edw
olf
(Can
isru
fus)
Hy
bri
diz
atio
nm
on
ito
rin
gA
dam
s&
Wai
ts(2
007)
Fae
ces
Mam
mal
sD
ho
le(C
uon
alpi
nu
s)P
op
ula
tio
ng
enet
ics
and
ph
ylo
geo
gra
ph
y
Iyen
gar
etal
.(20
05)
Fae
ces
Mam
mal
sB
row
nb
ear
(Urs
us
arct
os)
Po
pu
lati
on
size
esti
mat
ion
Bel
lem
ain
etal
.(20
05)
Fae
ces
Mam
mal
sW
ild
wes
tern
go
rill
a(G
oril
lago
rill
ago
rill
a)K
insh
ipas
soci
atio
ns
Bra
dle
yet
al.(
2007
)
Fae
ces
Mam
mal
sM
ult
iple
fur
seal
spec
ies
(Arc
toce
phal
us
sp)
Die
td
eter
min
atio
nC
asp
eret
al.(
2007
)
Fae
ces
Mam
mal
sW
olf
(Can
islu
pus)
Po
pu
lati
on
den
sity
Cre
elet
al.(
2003
)
Fae
ces
Mam
mal
sD
eser
tb
igh
orn
shee
p(O
vis
can
aden
sis
nel
son
i)G
ene
flo
wes
tim
atio
nE
pp
set
al.(
2006
)
Fae
ces
Mam
mal
sM
ult
iple
carn
ivo
resp
ecie
sD
iet
det
erm
inat
ion
Far
rell
etal
.(20
00)
Fae
ces
Mam
mal
sE
ura
sian
bad
ger
(Mel
esm
eles
)P
op
ula
tio
nsi
zees
tim
atio
nF
ran
tzet
al.(
2003
)
Fae
ces
Mam
mal
sA
tlan
tic
spo
tted
do
lph
ins
(Ste
nel
lafr
onta
lis)
Met
ho
do
log
yG
reen
etal
.(20
07)
Fae
ces
Mam
mal
sA
mu
rti
ger
(Pan
ther
ati
gris
alta
ica)
Ind
ivid
ual
iden
tifi
cati
on
by
scen
t-m
ark
ing
do
gs
Ker
ley
&S
alk
ina
(200
7)
Fae
ces
Mam
mal
sC
oy
ote
(Can
isla
tran
s)P
op
ula
tio
nsi
zees
tim
atio
nK
oh
net
al.(
1999
)
Fae
ces
Mam
mal
sO
tter
(Lu
tra
lutr
a)M
eth
od
olo
gy
Lam
pa
etal
.(20
08)
Fae
ces
Mam
mal
sB
igh
orn
shee
p(O
vis
can
aden
sis
can
aden
sis)
Ho
stg
enet
icd
iver
sity
and
par
asit
ism
Lu
ikar
tet
al.(
2008
)a
Fae
ces
Mam
mal
sC
him
pan
zee
(Pan
trog
lody
tes
veru
s)M
eth
od
olo
gy
Mo
rin
etal
.(20
01)
Fae
ces
Mam
mal
sIb
eria
nly
nx
(Lyn
xpa
rdin
us)
Sp
ecie
sid
enti
fica
tio
nP
alo
mar
eset
al.(
2002
)
Fae
ces
Mam
mal
sR
ock
wal
lab
y(P
etro
gale
pen
icil
lata
)P
op
ula
tio
nd
ensi
tyP
igg
ott
etal
.(20
06)
Fae
ces
Mam
mal
sL
esse
rh
ors
esh
oe
bat
(Rhi
nol
ophu
shi
ppos
ider
os)
Met
ho
do
log
yP
uec
hm
aill
eet
al.(
2007
)
Fae
ces
Mam
mal
sK
itF
ox
(Vu
lpes
mac
roti
sm
uti
ca)
Po
pu
lati
on
gen
etic
sS
mit
het
al.(
2005
)
Fae
ces
Mam
mal
sW
este
rng
ori
lla
(Gor
illa
gori
lla
gori
lla)
and
bar
bar
y
mac
aqu
e(M
acac
asy
lvan
us)
Met
ho
do
log
yV
alle
tet
al.(
2008
)
Fae
ces
and
hai
rM
amm
als
Eu
rop
ean
pin
em
arte
n(M
arte
sm
arte
s)an
dst
on
e
mar
ten
(Mar
tes
foin
a)
Sp
ecie
sid
enti
fica
tio
nR
uiz
-Go
nza
lez
etal
.(20
08)
� 2009 Blackwell Publishing Ltd
1280 T E C H N I C A L R E V I E W
Tab
le1
(Co
nti
nu
ed)
Gro
up
Sp
ecie
sP
urp
ose
Stu
dy
Fae
ces
and
hai
rM
amm
als
Bla
ckb
ear
(Urs
us
amer
ican
us)
Sp
atia
lb
ehav
iou
rS
chw
artz
etal
.(20
06)
Fae
ces
and
hai
rM
amm
als
Wo
lver
ine
(Gu
logu
lo)
Sp
ecie
san
din
div
idu
al
iden
tifi
cati
on
Uli
zio
etal
.(20
06)
Fae
ces
and
uri
ne
Mam
mal
sW
olv
erin
e(G
ulo
gulo
)In
div
idu
alid
enti
fica
tio
nan
d
gen
der
det
erm
inat
ion
Hed
mar
ket
al.(
2004
)
Fae
ces,
slo
ug
hed
skin
and
egg
shel
l
Am
ph
ibia
ns
and
rep
tile
s
Co
mm
on
Eu
rop
ean
vip
er(V
iper
abe
rus)
,ri
ng
edsn
ake
(Nat
rix
nat
rix)
and
smo
oth
snak
e(C
oron
ella
aust
riac
a)
Met
ho
do
log
yJo
nes
etal
.(20
08)
Fea
ther
sB
ird
sG
reat
erfl
amin
go
(Pho
enic
opte
rus
rose
us)
Gen
der
det
erm
inat
ion
Bal
kiz
etal
.(20
07)
Fea
ther
sB
ird
sP
ow
erfu
lo
wl
(Nin
oxst
ren
ua)
Met
ho
do
log
yH
og
anet
al.(
2008
)
Fea
ther
sB
ird
sL
esse
rsp
ott
edea
gle
(Aqu
ila
pom
arin
a)S
oci
alo
rgan
izat
ion
Mey
bu
rget
al.(
2007
)
Fea
ther
sB
ird
sE
aste
rnim
per
ial
eag
le(A
quil
ahe
liac
a)S
pec
ies
iden
tifi
cati
on
Ru
dn
ick
etal
.(20
07)
Fea
ther
sB
ird
sE
aste
rnim
per
ial
eag
le(A
quil
ahe
liac
a)P
op
ula
tio
ng
enet
ics
Ru
dn
ick
etal
.(20
08)
Fea
ther
sB
ird
sC
aper
cail
lie
(Tet
rao
uro
gall
us)
Met
ho
do
log
yS
egel
bac
her
(200
2)
Fea
ther
s(i
ncl
ud
ing
mu
seu
msp
ecim
ens)
Bir
ds
Sp
anis
him
per
ial
eag
le(A
quil
aad
albe
rti)
Met
ho
do
log
yH
orv
ath
etal
.(20
05)
Fea
ther
san
deg
gsh
ell
Bir
ds
47b
ird
spec
ies
Gen
der
det
erm
inat
ion
Jen
sen
etal
.(20
03)
Fo
ot
mu
cus
Inv
erte
bra
tes
Mu
ltip
lete
rres
tria
lsn
ails
Met
ho
do
log
yP
alm
eret
al.(
2008
)
Fo
ot
mu
cus
Inv
erte
bra
tes
Mu
ltip
lein
tert
idal
snai
lsM
eth
od
olo
gy
Kaw
aiet
al.(
2004
)
Fre
shw
ater
Am
ph
ibia
ns
and
rep
tile
s
Am
eric
anb
ull
fro
g(R
ana
cate
sbei
ana)
Sp
ecie
sid
enti
fica
tio
nF
icet
ola
etal
.(20
08)
Hai
rM
amm
als
Do
mes
tic
do
g(C
anis
fam
ilia
ris)
Met
ho
do
log
yB
jorn
erfe
ldt
&V
ila
(200
7)
Hai
rM
amm
als
San
Joaq
uin
kit
fox
(Vu
lpes
mac
roti
sm
uti
ca)
Met
ho
do
log
yB
rem
ner
-Har
riso
net
al.(
2006
)
Hai
rM
amm
als
Bla
ckb
ear
(Urs
us
amer
ican
us)
Po
pu
lati
on
den
sity
Dre
her
etal
.(20
07)
Hai
rM
amm
als
Gia
nt
pan
da
(Ail
uro
poda
mel
anol
euca
)G
end
erd
eter
min
atio
nD
urn
inet
al.(
2007
)
Hai
rM
amm
als
Ora
ng
-uta
n(P
ongo
spp
)M
eth
od
olo
gy
Go
oss
ens
etal
.(20
04)
Hai
rM
amm
als
Mu
ltip
leN
ort
hA
mer
ican
carn
ivo
res
Po
pu
lati
on
gen
etic
sK
end
all
&M
cKel
vey
(200
8)
Hai
rM
amm
als
Mo
un
tain
py
gm
y-p
oss
um
(Bu
rram
yspa
rvu
s)G
enet
icd
iver
sity
Mit
rov
ski
etal
.(20
07)
Hai
rM
amm
als
Bro
wn
bea
r(U
rsu
sar
ctos
)In
div
idu
als
abu
nd
ance
Mo
wat
&P
aetk
au(2
002)
Hai
rM
amm
als
Eu
rasi
anly
nx
(Lyn
xly
nx)
Po
pu
lati
on
mo
nit
ori
ng
Sch
mid
t&
Ko
wal
czy
k(2
006)
Hai
rM
amm
als
So
uth
ern
hai
ry-n
ose
dw
om
bat
(Las
iorh
inu
sla
tifr
ons)
Sp
atia
ld
istr
ibu
tio
nan
d
hab
itat
use
Wal
ker
etal
.(20
08)
Hai
rM
amm
als
Oce
lot
(Leo
pard
us
pard
alis
)S
pec
ies,
gen
der
ind
ivid
ual
iden
tifi
cati
on
Wea
ver
etal
.(20
05)
Hai
rM
amm
als
Mu
ltip
leca
rniv
ore
spec
ies
Met
ho
do
log
yZ
ieli
nsk
iet
al.(
2006
)
Hai
r⁄f
aece
s⁄
uri
ne
⁄to
oth
⁄sal
iva
Mam
mal
sW
olf
(Can
islu
pus)
Gen
der
det
erm
inat
ion
Sas
tre
etal
.(20
08)
Inse
ctex
uv
iae
⁄fra
ssIn
ver
teb
rate
sM
ult
iple
bu
tter
fly
spec
ies
Sp
ecie
sid
enti
fica
tio
nF
ein
stei
n(2
004)
Ivo
ryM
amm
als
Afr
ican
Ele
ph
ant
(Lox
odon
taaf
rica
na
spp)
Fo
ren
sic
case
sW
asse
ret
al.(
2007
)
Men
stru
alb
leed
ing
Mam
mal
sT
aiw
anm
acaq
ue
(Mac
aca
cycl
opis
)M
eth
od
olo
gy
Ch
uet
al.(
1999
)
Mu
seu
msp
ecim
enB
ird
sG
alli
nag
osp
pM
eth
od
olo
gy
Lee
&P
rys-
Jon
es(2
008)
� 2009 Blackwell Publishing Ltd
T E C H N I C A L R E V I E W 1281
improved techniques that can be applied in noninvasive
genetics to improve both data production and analysis.
Unfortunately, these scientific communities seldom
cross-reference each other. To continually improve
molecular ecology and conservation genetic studies, we
recommend that researchers occasionally search for
novel approaches in journals from diverse fields includ-
ing forensics (e.g. Journal of Forensic Sciences), human and
animal health (Avian Disease, New England Journal of Medi-
cine), microbiology (e.g. Journal of Applied Microbiology),
biochemistry and biotechniques (Analytical Biochemistry
and Nature Methods), and bioinformatics, e.g. Biometrika
(see also our literature cited).
This review is structured around the steps in a nonin-
vasive study, from pre-PCR sampling to post-PCR data
analysis, and concludes with perspectives for future
research. Noninvasive studies should not be seen as a
one-step process, but as a chain of steps that should be
monitored independently. The chain starts in the living
animal and ends only when the statistical analyses of the
final data provide convincing evidence that results and
conclusions are reliable. We consider five major steps to
be monitored and how to avoid pitfalls and improve non-
invasive studies (Fig. 1). Accordingly, this review is
structured around steps and techniques, not research
questions (e.g. paternity analysis, population structure),
which allows readers to quickly go to the step or
technique of interest (pre-PCR to post-PCR) to find
information.
Pre-PCR
Obtaining samples
Creative ways to noninvasively obtain DNA from
numerous types of samples are continuously being
developed, improved and evaluated (Table 1). The
collection of everything from menstrual fluid to mucus
trails left by snails has been used to identify species and
individuals noninvasively (Table 1). Several sample
types can be obtained by following a trail of an animal on
natural surfaces such as snow or sand, without ever
seeing the target animals. For example, Ulizio et al. (2006)
collected 169 hair samples and 58 scat samples on 54
wolverine backtracks.
One creative study reported the noninvasive detection
of species (a frog, Rana catesbeiana) in natural wetlands by
PCR testing for mtDNA in water samples (Ficetola et al.
2008).
Faeces are one of the most commonly used noninva-
sive materials because, for many species, it is the easiest
to find in the wild and it provides more information (e.g.
diet, stress hormone status, reproductive hormones, par-
asite infection and parasite DNA) than other sampleTab
le1
(Co
nti
nu
ed)
Gro
up
Sp
ecie
sP
urp
ose
Stu
dy
Mu
seu
msp
ecim
enM
amm
als
Bro
wn
bea
r(U
rsu
sar
ctos
)P
hy
log
eog
rap
hy
Leo
nar
det
al.(
2000
)
Mu
seu
msp
ecim
enM
amm
als
Sto
at(M
ust
ela
erm
inea
)M
eth
od
olo
gy
Mar
tin
ko
va
&S
earl
e(2
006)
Mu
seu
msp
ecim
enM
amm
als
Wo
lver
ine
(Gu
logu
lo)
Ev
olu
tio
nar
ysi
gn
ifica
nt
un
its
Sch
war
tzet
al.(
2007
)
Reg
urg
itat
eM
amm
als
Wo
lf(C
anis
lupu
s)In
div
idu
als
dis
per
sio
nV
alie
re&
Tab
erle
t(2
000)
Sal
iva
Bir
ds
Co
mm
on
mar
mo
set
(Cal
lith
rix
jacc
hus)
Co
rtis
ol
lev
els
and
beh
avio
ura
lst
ress
Cro
sset
al.(
2004
)
Sal
iva
Mam
mal
sW
ild
chim
pan
zee
(Pan
trog
lody
tes
veru
s)In
div
idu
alid
enti
fica
tio
nIn
ou
eet
al.(
2007
)
Sal
iva
Mam
mal
sW
olf
(Can
islu
pus)
Pre
dat
or
iden
tifi
cati
on
Su
nd
qv
ist
etal
.(20
08)
Sal
iva
Mam
mal
sC
oy
ote
(Can
isla
tran
s)P
red
ato
rid
enti
fica
tio
nB
lejw
aset
al.(
2006
)
Sce
nt
mar
kM
amm
als
Mu
ltip
lem
uri
ne
spec
ies
Mic
rob
ial
par
asit
eco
mm
un
itie
sid
enti
fica
tio
nL
any
on
etal
.(20
07)
Sk
in,b
lub
ber
and
mea
tM
amm
als
Pac
ific
min
ke
wh
ale
(Bal
aen
opte
raac
uto
rost
rata
spp
)F
ore
nsi
cca
ses
Bak
eret
al.(
2007
)
Slo
ug
hed
⁄sh
edsk
inM
amm
als
Hu
mp
bac
kw
hal
e(M
egap
tera
nov
aean
glia
e)M
eth
od
olo
gy
Elp
hin
sto
ne
etal
.(20
03)
Slo
ug
hed
⁄sh
edsk
inM
amm
als
Hu
mp
bac
kw
hal
e(M
egap
tera
nov
aean
glia
e)In
div
idu
als
abu
nd
ance
Pal
sbo
llet
al.(
1997
)
Slo
ug
hed
⁄sh
edsk
inM
amm
als
Rin
ged
seal
(Pho
cahi
spid
a)M
eth
od
olo
gy
Sw
anso
net
al.(
2006
)
Uri
ne
Mam
mal
sJa
pan
ese
mac
aqu
es(M
acac
afu
scat
a)M
eth
od
olo
gy
Hay
akaw
a&
Tak
enak
a(1
999)
Uri
ne
Mam
mal
sW
olv
erin
e(G
ulo
gulo
)M
eth
od
olo
gy
Hed
mar
ket
al.(
2004
)
Uri
ne
Mam
mal
sW
olf
(Can
islu
pus)
Po
pu
lati
on
mo
nit
ori
ng
Hau
skn
ech
tet
al.(
2007
)
Uri
ne
Mam
mal
sM
ult
iple
can
idsp
ecie
sS
pec
ies
and
ind
ivid
ual
iden
tifi
cati
on
Val
iere
&T
aber
let
(200
0)
Uri
ne
Mam
mal
sW
olf
(Can
islu
pus)
Mo
lecu
lar
sex
ing
Sas
tre
etal
.(20
08)
� 2009 Blackwell Publishing Ltd
1282 T E C H N I C A L R E V I E W
types (Kohn & Wayne 1997; Goymann 2005; Luikart et al.
2008a; Schwartz & Monfort 2008). Faeces in some species
(e.g. ungulates, arboreal primates, macropods, etc.) can
be collected just after observing individuals defaecate
without disturbing the animals. An enormous advantage
of observing the target animals is that the faeces are fresh
and DNA is relatively little degraded. It can also help
avoid collecting from nontarget species and determine
sex (by observation) and thus avoid need for DNA-based
species identification and sexing (Epps et al. 2006; Luikart
et al. 2008b). In secretive or elusive species such as forest
ungulates, bears, fishers, mountain lions and tigers,
domestic dogs can be trained to find faeces (reviewed in
McKay et al. 2008). Dogs can also identify individual
animals, as was shown in a study of faeces from known
tigers (Kerley & Salkina 2007). Hair is another widely
collected material (Table 1). In apes (e.g. orangutans,
chimpanzees), individuals build a new nest every night
and hair that is shed during the night can be found in the
nests. Researchers recommend using only hairs with
visible root bulbs as many shed hairs do not contain large
bulbs with DNA (Goossens et al. 2004). In a study of
wolves, hair (along with faeces, urine and saliva) allowed
highly successful DNA amplification (93% of samples)
for noninvasive sexing of individuals using sex
chromosome markers (Sastre et al. 2008). Hair is also
often recovered frozen in the snow tracks of felids and
canids and in bed sites of ungulates.
Many hair snare devices have been invented (e.g.
Bremner-Harrison et al. 2006; Zielinski et al. 2006) for
noninvasive sampling. Hair snares are used to sample
bears (e.g. Immell & Anthony 2008; Kendall et al. 2009),
felids (e.g. Weaver et al. 2005; Schmidt & Kowalczyk
2006) and mustelids Mowat & Paetkau 2002. Barbed wire
or sticky tape is also often strung around bait stations or
draped across animal burrow entrances to pluck hairs
when animals pass by (Pauli et al. 2008; Toth 2008;
Walker et al. 2008). Along with hair snaring devices, com-
mercial lures (such as catnip and valerian oils, among
other attractants) have been successfully used to attract
and elicit cheek-rubbing behaviour in felid species (e.g.
Lynx canadensis, McDaniel et al. 2000). For hair snares, a
potential advantage is that they obtain plucked hairs,
which generally contain more and larger root bulbs (with
cells and DNA) than shed hairs. However, it might be
difficult avoiding cross-contamination between individu-
als because multiple individuals can be sampled before
hairs are recovered from the snare. As birds use mammal
hair to strengthen the structure of their nest, recently
Toth (2008) used bird nests as sources of hair samples
and identify mammals that occupy or migrate through a
specific area.
Feathers have repeatedly been shown to be a good
source of DNA. Shed feathers can be collected from nests.
Feather snares (e.g. sticky tape) potentially could help
obtain feather samples, but to our knowledge have not
been reported in the literature. A particularly informative
and recent study (Hogan et al. 2008) showed that differ-
ent feather types (down, semi plume, contour or remi-
ge ⁄ rectrice) yield useful DNA. However, feather
Fig. 1 Schematic representation of some critical points (light grey lists) that should be checked at each of the five main steps (white
rectangles on top) of the noninvasive samples processing. Below (arrow boxes) are some likely consequences of not correctly following
and monitoring these points. Some points are common sense and widely known but nonetheless are often violated. HWE, Hardy–
Weinberg equilibrium proportions; LD, linkage- or genotypic-disequilibrium.
� 2009 Blackwell Publishing Ltd
T E C H N I C A L R E V I E W 1283
condition (as estimated from physical appearance)
strongly influenced PCR amplification success.
For eggshells, a recent study used cotton swabs to
obtain DNA from the external shells of herring gull (Larus
argentatus) and Caspian tern (Sterna caspia) eggs (Handel
et al. 2006). Researchers verified that the DNA samples
were maternal (not the chick’s) by comparing microsatel-
lite profiles with those obtained from adults and chicks
from the same nests. In all of 28 tests, the egg swabs
matched the maternal microsatellite genotype. In a screen-
ing of many nests of both species, microsatellite markers
were successfully amplified from egg swabs. Eggshells
are also used as a source of DNA in veterinary health and
disease studies (e.g. Miller et al. 2003), from which molec-
ular ecologists might learn new and useful techniques
(e.g. improved DNA extraction or PCR techniques).
Eggshells, feathers and mouth swabs from sage-
grouse (Centrocercus urophasianus) were compared for
PCR success in one of the few studies directly comparing
multiple sample types (Bush et al. 2005). These authors
found hatched eggshell membranes yield useful DNA
(better than predated eggshells), as did plucked body
contour feathers, chick down feathers and mouth swabs.
However, allelic dropout rates of approximately 10%
were observed for eggshells, and moulted feathers had
only 60% PCR amplification success (Bush et al. 2005).
Saliva is also a good source of DNA. It is often used in
forensics, for example, to recover DNA from bite marks
found in homicides, assault and other criminal cases
(Anzai-Kanto et al. 2005). In wildlife, Williams et al. (2003)
used saliva collected from sheep bite wounds to identify
the canid species responsible for attacks on domestic
sheep; the authors identified the predator species (coyote)
and determined the sex of the individual. Saliva is also
used to solve cases of livestock attacks in which wolves
and dogs are the main suspects (Sundqvist et al. 2008).
For some sample types, a new swab sampling
technique reported in the forensics literature could
improve the quality of genotyping Pang & Cheung, 2007.
The double swab technique, using a wet cotton swab
followed by a dry cotton swab, was compared with the
classical technique (one wet swab) for recovering DNA
from evidence collected at crime scenes. Swab techniques
could potentially improve noninvasive sampling studies
involving material such as eggshells or any surfaces that
animals come into contact, rub against, lick or bite (e.g.
rocks, sticks). Further evaluation of this and other
sampling methods is needed.
Preserving DNA in noninvasive samples
A growing diversity of protocols exists for preserving
DNA in samples. This makes it difficult for researchers to
understand which protocol is most reliable, most
thoroughly validated, or requires further development
and testing. For noninvasive samples, it is essential to
conduct a pilot study using the exact target material,
preservation method and extraction technique to ensure
recovery of sufficient DNA (Bhagavatula & Singh 2006;
Valiere et al. 2006; Schwartz & Monfort 2008).
The preservation of DNA in a noninvasive sample is a
race to inhibit enzymes that degrade DNA, i.e. nucleases.
There are three main approaches used to preserve sam-
ples: deactivation of nucleases via removal of water,
deactivation of nucleases via the elimination of cations
(e.g. MgCl2; Thomas & Gilbert 2006) and inhibition of
nuclease activity via storage of samples at low tempera-
tures. Removal of water is achieved using drying agents
(e.g. ethanol, silica gel) or drying techniques (e.g. vacuum
spinning, lyophilization, oven heating). Removal of
cations is achieved using chelators such as EDTA or resin
(e.g. Chelex�). Insufficient volumes of preservatives (eth-
anol or silica) or failure to freeze samples quickly often
leads to DNA degradation. Several published studies
comparing different preservation protocols can help
researchers choose the best protocol according to their
samples (Roon et al. 2003; Hajkova et al. 2006; Broquet
et al. 2007b; Santini et al. 2007). Nonetheless, there are
inconsistencies among some studies, and even some sug-
gestion that there is an interaction between preservation
techniques and extraction methods (Piggott & Taylor
2003).
A potentially improved preservation approach is to
combine use of silica and ethanol (ETOH) protocols,
however, the ETOH (90%) performed similar to the com-
bined two-step method when using lower quality sam-
ples (Roeder et al. 2004). This combined approach was
repeated on gorilla and chimpanzee faeces and yielded
more DNA than silica alone or RNAlater alone (Nsubuga
et al. 2004). Nonetheless, this approach has not been
extensively compared with other methods (e.g. ETOH
97%) in a wide variety of species. Long-term preservation
might be improved by the addition of trehalose as a pre-
servative agent (Smith & Morin 2005), although this
method has not been independently evaluated. There is a
great need for comparative evaluations of most preserva-
tion methods.
For faeces preservation, it is difficult to decide which
desiccant (e.g. ETOH, silica, salts) should be used.
A large amount of any desiccant should be used per
sample (e.g. 5–10 parts of desiccant per part of sample) to
rapidly and completely dry the sample material. Given
the wide use and success, we recommend the use of
ETOH in large volumes (5–10 times the sample volume)
and in high concentration (‡95% ETOH).
With faecal material, ETOH has advantages over silica
in that ETOH prevents formation of faecal powders (thus
cross-contamination by aerosol). It also keeps the external
� 2009 Blackwell Publishing Ltd
1284 T E C H N I C A L R E V I E W
mucous layer containing cells packed against faecal
material, whereas silica can be abrasive and can remove
mucus and cells from outer surface of the faeces (e.g. dur-
ing transportation and shaking of samples). ETOH has a
notable disadvantage in being flammable and therefore
potentially dangerous and more expensive to ship via
airplane. As an alternative, silica is useful and widely
tested but again requires large volumes of this mineral to
ensure rapid drying and to avoid exhausting the desicca-
tion function.
For faeces, RNA later� might be a better preservative
than ETOH or silica http://www.aim.uzh.ch/orangu-
tannetwork/GeneticSamplingProtocol.html#18; Nsubuga
et al. 2004). RNA later� is a solution meant for preserving
RNA in tissue. However, the solution is expensive
(US$2–4 depending on volume needed per sample), and
further research is needed to formally test and compare it
with other preservatives.
For hair samples, the most common storage method is
simply to store it (shed or plucked) in a dry envelope
often with silica gel granules at room temperature (Jeff-
ery et al. 2007). A fairly thorough comparative study of
freezing ()20 �C) vs. silica desiccant found that freezing
gives slightly higher (though nonsignificant) amplifica-
tion success for both microsatellites and mtDNA from
brown bear hair, Ursus arctos (Roon et al. 2003). Amplifi-
cation success was above approximately 90% up to
6 months of storage but dropped below approximately
80% between 6 and 12 months for both the 1000-bp
mtDNA fragment and three microsatellite loci. More
comparative studies are needed using different preserva-
tion methods including a combination of freezing and
silica gel, and perhaps storing hairs immediately into a
lysis or storage buffer solution. Sorting hairs based on
root bulb size and quality should also be conducted to
maximize amplification success and perhaps improve
accuracy of comparisons among preservation methods
(Jeffery et al. 2007).
For feathers, storage in paper envelopes at )20�allowed successful amplification of mtDNA and nDNA
from powerful owls (Ninox strenua; Hogan et al. 2008). In
this study, the paper envelopes containing 637 shed
feathers were stored in plastic bags in dry and dark con-
ditions for up to 7 months. Amplification success was
80–90% for mtDNA and microsatellites on feathers in
good condition but only 30–40% for feathers in poor con-
dition (with visible physical degradation of calamus and
barbs on the vane). Feather type had no effect on amplifi-
cation success. We recommend against using plastic bags
as humidity can potentially build up inside, unless silica
desiccant is inside the bag. Feathers from adult eagles
(Rudnick et al. 2007) stored dry at room temperature
yielded microsatellite genotypes using a pre-amplifica-
tion PCR method (PCR section below), although nearly
10% of samples yielded no PCR product. In the same
study, developing chick feathers were stored at room
temperature in a lyses buffer (EDTA, SDS) for several
months before being ultimately stored at )80 �C up to
several years before yielding microsatellite genotypes.
Saliva samples are generally preserved by freezing at
)20 �C (Anzai-Kanto et al. 2005). For example, Anzai-
Kanto et al. (2005) published a study using human saliva
in which they estimate that 0.3 mL of saliva is enough to
provide DNA for genotyping 15 loci. Swabs are the most
general method to sample buccal ⁄ oral DNA, and these
swabs are generally dried at room temperature followed
by freezing at )20 �C or even colder temperatures (e.g.
see Sundqvist et al. 2008).
Urine samples as a source of DNA have been increas-
ingly used in recent years. Urine can be collected using a
swab to swipe the surface location where the animal uri-
nated (e.g. rocks, sticks, leaves). The swab will absorb the
urine together with the cells. Another method, used in
winter, is the collection of urine in snow (yellow snow).
Researchers have melted yellow snow in a 15-mL tube,
which will contain urine, cells and DNA (Hausknecht
et al. 2007). This method has been tested in carnivores, in
particular the wolf. Urine samples can also be collected
from soil samples. We have collected fresh ungulate
urine from dirt, which becomes mud (G. Luikart, unpub-
lished). We stored the urine mud in six volumes of 95%
ETOH, similar to faecal samples, until extraction in the
laboratory using stool extraction kits or soil kits (see
below). While urine can be a useful material, it often has
a lower amplification success rate as compared with
other noninvasive samples (Hedmark et al. 2004).
Hedmark et al. noticed a decline in microsatellite amplifi-
cation success of wolverine urine (40% success) as com-
pared with faeces (65% success).
Extracting DNA from noninvasive samples
DNA extraction is a crucial step, because all subsequent
steps in a genetic project hinge upon extraction quality.
Phenol ⁄ chloroform extraction methods were the most
widely used 10–15 years ago, but now are seldom used,
mostly because the chemicals are hazardous, the
approach is time-consuming, and sometimes PCR inhibi-
tors remain after extraction. As alternatives, different
methods have appeared, most of them imported from
forensic genetics (e.g. see book by Morling 2008).
Resin-based (e.g. Chelex�) extractions are widely used
for noninvasively collected samples. Chelex is useful for
extracting DNA from hair follicles (Mitrovski et al. 2005;
Koukoulas et al. 2008), stains at crime scenes, and even
for formalin-fixed archived tissues (Chakraborty et al.
2006). Its main advantages are speed and low cost
(http://bugs.bio.usyd.edu.au/DNA/DNAextrn.html).
� 2009 Blackwell Publishing Ltd
T E C H N I C A L R E V I E W 1285
The main disadvantages are that (i) DNA extracts are not
always highly pure, (ii) DNA can degrade after several
months, and (iii) Chelex itself can inhibit the PCR ampli-
fication (Willard et al. 1998).
Commercial kits for extracting DNA are also widely
used. Among these, the most common are silica-based
spin column kits. The working principle of this method
involves the lysis of the cell membranes (e.g. by deter-
gents and proteinase K), followed by purification using
silica-based compounds in spin columns that bind and
then allow washing of DNA (Boom et al. 1990). The great
success of these kits results from their ease of use and
adaptability to a wide range of biological samples (e.g.
plant tissues, bacteria growing media, skin, muscle, bone,
faeces, urine, blood, museum skins, ancient bone) with
minimum changes.
When comparing five DNA extraction methods, the
extracted samples from which a fragment of 149 bp of
the mtDNA was successfully PCR amplified using a com-
mercial kit (QIAGEN Stool DNA extraction kit) was
100%, followed by 88% using guanidinium thiocyanate-
silica, 75% for the digest buffer ⁄ phenol–chloroform, 38%
for chelex-100 and 25% for the lyses buffer ⁄ column puri-
fication method (Bhagavatula & Singh 2006)
For pellet-form faeces, which are amenable to a sur-
face wash, the wash technique combined with commer-
cial extraction kits [e.g. DNeasyTM Blood Kit (QIAGEN)]
has been highly successful. The washing step is a simple
10–15 min incubation of a faecal pellet in a buffer solu-
tion followed by extraction of DNA from the buffer using
a blood DNA extraction kit (Luikart et al. 2008b). The
surface-wash liquid contains relatively few PCR inhibi-
tors and therefore does not always require use of the
more expensive and time-consuming ‘stool kits’ with
additional steps to remove inhibitors. This approach
yields high amplification success, low genotyping error
rates and large quantities of DNA.
For faeces, a cell enrichment method has been
reported to recover large quantities of high molecular
weight DNA (Wan et al. 2006). The cell enrichment based
protocol is so far the only one that deals with large quan-
tity of faeces, and is based on the soaking in a large vol-
ume of buffer to disperse the faecal material completely.
A commercial company (Noninvasive Technologies)
offers a kit for a similar extraction, but it costs over
US$200 for the extraction of two individual samples.
With faecal (& urine) samples, it is difficult to quantify
the amount of extracted DNA using conventional meth-
ods (e.g. spectrophotometer) because these are inefficient
with trace quantities of DNA, they cannot estimate
DNA degradation, nor can they differentiate between
DNA from the target species or microbes often in faecal
(& urine) DNA extractions. To cope with these
limitations, several assays have been developed using
real-time quantitative (RTQ) PCR (Morin et al. 2007).
Unfortunately, RTQ-PCR still is not affordable for all lab-
oratories and alternative low-cost methods can be used
to quantify the DNA extracted from some noninvasive
samples. For example, Ball et al. (2007) used a method
based on PicogreenTM (Molecular probes), a fluorescent
dye, to measure the amount of double-stranded DNA
extracted from noninvasive samples (e.g. faeces). Pico-
greenTM binds double-stranded DNA and when excited
by laser releases a florescent signal that is proportional
to the amount of double-stranded DNA present in the
tested aliquot. However, unlike RTQ-PCR, fluorescent
dye methods cannot differentiate between the target
species vs. microbial DNA.
The urine samples can be collected either by using a
swab across the surface where the animal urinated (e.g.
rocks, bush leafs) or in winter from snow. One extraction
method involves centrifuging cells (sloughed off from
the epithelium of the urinary tract). Once the cells are
collected in a pellet, standard DNA extraction protocols
can be used (Hausknecht et al. 2007). This approach is
also valid for buccal-mouth wash (nondestructive) sam-
pling in humans (Mayntz-Press & Ballantyne 2007).
Another extraction method directly precipitates DNA
from the sample (e.g. snow) containing the urine (Vali-
ere & Taberlet 2000). Direct precipitation would be
advantageous when cells burst and DNA is free. DNA
from the urine deposited in the soil (mud) can be
obtained using stool DNA extraction kits or soil DNA
extraction kits (e.g. Thakuria et al. 2009). Comparative
evaluations of extraction kits on humans suggest that
some kits (miniMAG) yield far better DNA than others,
including DNA from pathogens being monitored nonin-
vasively (Tang et al. 2005). Noninvasive wildlife studies
might benefit from testing and using kits used in human
studies.
For hairs, an improved extraction method reported
use of Ca+ to increase digestion and release of DNA of
hair shafts. In a forensic-based study of hairs from 170
dogs from different breeds, the quantity of DNA
extracted increased 100% when compared to the well-
established QIAGEN tissue kit (Pfeiffer et al. 2004).
Finally, it is important to mention that plastic tubes
may have a strong effect of reducing DNA quantities
when the amount of DNA in the sample is very low
(fewer than 1000 target copies) because of DNA adhering
to the plastic walls of the tube. A recent study showed
that use of low-retention plastic tubes significantly
reduce DNA loss, but DNA from nontarget species
added to prevent the loss of target DNA had no effects
(Ellison et al. 2006). As this problem becomes better
understood, we imagine that low-retention plastic tubes
will drop in price; more research is needed on changes in
DNA yield caused by tube choice.
� 2009 Blackwell Publishing Ltd
1286 T E C H N I C A L R E V I E W
Polymerase chain reaction
Here we review approaches to improve PCR amplifica-
tion of DNA, including pre-PCR treatments (for inhibi-
tors and broken DNA fragments), amplification of
smaller fragments (mini-STRs ⁄ microsatellites and SNPs),
nested PCR techniques, different Taq polymerase
enzymes and genotype scoring criteria.
Overcoming PCR inhibitors
Inhibition of PCR can cause low amplification rates, even
in samples with abundant DNA and apparently suitable
for PCR (Kontanis & Reed 2006). For example, faeces con-
tain compounds that can be strong PCR inhibitors,
including complex polysaccharides, products from food
degradation (e.g. acids, secondary plant compounds,
enzymes, lipids and proteins), RNA and bacteria. As pre-
viously discussed, DNA extraction protocols combined
with washes for DNA purification are essential to remove
inhibitors. However, some inhibitors may still remain
and result in amplification failure.
Dilution of the DNA extracts is the simplest way to
reduce inhibitors (dilution is the solution to pollution).
For example, Thornton & Passen (2004) diluted approxi-
mately 256-fold the DNA extract obtained from 10 mg of
bovine faeces to achieve amplification inhibited by phytic
acid (present in plants). Dilution also increased amplifi-
cation efficiency of Iberian lynx (Lynx pardinus) mtDNA
from 92.6% to 99%, equivalent to the benefit of perform-
ing a second amplification for each sample (Palomares
et al. 2002). However, genotyping errors can be caused by
low target DNA quantity or the presence of PCR inhibi-
tors (or both interacting). Accordingly, a balance between
diluting PCR inhibitors and over-diluting the DNA in the
extract often must be established.
Precipitation of DNA (e.g. with ETOH) also removes
inhibitors (and increase DNA concentration). This
involves a washing step of the DNA pellet before re-dis-
solving the DNA precipitant in water or buffer. Addition
of PCR adjuvants such as bovine serum albumin (BSA),
dimethyl sulfoxide, or nonionic detergents (e.g. Tween 20
and Triton X-100) often binds inhibitors and improves
amplification specificity. Most noninvasive studies
include an additive in PCR protocols. BSA is the most
widely used adjuvant (from 0.1 to 1.2 lg ⁄ lL in concen-
tration) because it seldom interferes with PCR in the
absence of an inhibitor.
Overcoming DNA degradation and fragmentation
Using very short fragments such as mini short tandem
repeats (mini-STRs, also called mini-microsatellies) or
single nucleotide polymorphisms (SNPs) can help over-
come difficulties amplifying degraded DNA (e.g. Camp-
bell & Narum 2008). In several noninvasive studies, long
amplicons (>200–300 bp) produced significantly higher
allelic dropout rates than short amplicons (Broquet &
Petit 2004; Buchan et al. 2005). Several studies have rede-
signed primers to produce shorter amplicons and
improve microsatellite analysis in forensic research (e.g.
Butler et al. 2003; Chung et al. 2004). In fact, studies using
historical or ancient DNA typically amplify multiple
small (100 bp) regions, instead of one large region as is
typical with high-quality DNA (Schwartz et al. 2007).
Single nucleotide polymorphism studies can achieve
higher amplification success and lower error rates than
microsatellites, because SNP amplicons are generally
shorter (<100 bp) than microsatellite amplicons (100–
300 bp). For example, Musgrave-Brown et al. (2007)
showed that a 52-plex SNP assay performed better than
STR (microsatellite) typing on degraded samples. How-
ever, the biallelic nature (and thus limited heterozygos-
ity) of SNPs must be compensated by typing a larger
number of SNP loci (Morin et al. 2004, 2009a, b). Thus,
even if there is a lower error rate per SNP, the amplifica-
tion of many more SNPs may cumulatively increase the
overall (multilocus) genotyping error rates. More
research is needed to quantify the increase in multilocus
error rates when adding more loci because the increase
can be unpredictable given that errors are often not ran-
domly distributed among PCRs, alleles and loci (Pomp-
anon et al. 2005).
The benefit of amplifying shorter SNP fragments is
likely to outweigh the lower variation and need to
include more loci when using SNPs. For example, Camp-
bell & Narum (2008) genotyped chinook salmon samples
of varying quality with 13 microsatellite and 29 SNP
assays and the average genotyping success for good,
intermediate and poor quality samples was 98%, 97%
and 79% for SNPs but only 96%, 24% and 24% for micro-
satellite loci respectively. Few studies have quantified
genotyping error rate using SNPs in noninvasive or his-
torical samples. Morin & Mccarthy (2007) used 19 SNPs
in a study using historical samples of bowhead whales;
they found a 0.1% genotyping error rate, which is lower
than most noninvasive studies.
During PCR, broken DNA fragments may anneal to
each other and form priming sites needed for amplifica-
tion, resulting in different sized fragments and the scor-
ing of false alleles. To prevent this unwanted production
of chimeric alleles (e.g. DNA fragments that anneal
together giving the appearance of another allele) and to
avoid the occurrence of jumping PCR (recombination
between similar DNA sequences during PCR that is pro-
moted in damaged ⁄ fragmented DNA), Culjkovic et al.
(2003) described a pretreatment of DNA fragments before
PCR by adding a poly(A) tail at the 3¢ prime end of
� 2009 Blackwell Publishing Ltd
T E C H N I C A L R E V I E W 1287
templates to eliminate homology between fragments.
This has been successfully used in ancient DNA studies,
but not to our knowledge in noninvasive studies.
Overcoming low DNA quantity
Several PCR-based strategies to overcome problems asso-
ciated with low-quantity DNA have been proposed
recently. Pre-amplification (i.e. double amplification) is
an efficient procedure to increase the amount of low copy
number template because products from a first amplifica-
tion are used as templates for a subsequent PCR; this pre-
amplification increases the DNA available for the second
desired amplification (e.g. Lau et al. 2003). A second PCR
with internal (nested) primers can also increase genotyp-
ing success and specificity to amplify only the target
locus because the internal primers (as well initial external
primers) can be locus specific. The same is true when
using only one internal primer in the second PCR (Belle-
main & Taberlet 2004). A semi-nested or second PCR can
be especially useful to improve amplification of certain
difficult loci. A second PCR is also useful after whole
genome pre-amplification or multiplex pre-amplification.
Whole genome amplification is the production of
amplicons across an entire genome to increase the
amount of template DNA available for subsequent locus-
specific genotyping (Kittler et al. 2002). This approach
has been successfully applied before genotyping micro-
satellites, although preferential amplification of the
shorter alleles might occur. Similarly, whole-genome
amplification with degenerate primers (i.e. mixtures of
similar, but not identical, primers) has been successfully
used for large-scale SNP genotyping despite a detectable
loss in genotype accuracy (Grant et al. 2002). In some
studies, as the one reported by Vigilant (1999) in geno-
typing shed chimpanzee hairs, this strategy was ineffec-
tive for improving microsatellite genotyping.
Pre-amplification of multiple loci in a multiplex can
improve microsatellite genotyping from noninvasive
samples (Box 1). This method can increase the quantity
of target DNA fragments for each locus while minimizing
consumption of the initial DNA extract. In this approach,
Box 1. The promise of real-time quantitative PCR
Real-time quantitative PCR quantifies the amount of target-specific, ‘amplifiable’ DNA from an extraction. This is
important because DNA might exist in a sample (e.g. quantified by fluorometry), but not be amplifiable because of
PCR inhibitors, extreme DNA fragmentation, and ⁄ or the DNA is from nontarget species. RTQ-PCR differs from regu-
lar PCR in that the PCR product is quantified as the PCR is occurring, using a fluorescent dye. In each PCR cycle, the
amount of the target locus DNA doubles and so does the fluoresce intensity. An RTQ-PCR machine is a PCR machine
with a fluorometer. Advantages of RTQ-PCR are its sensitivity (it is the most sensitive PCR method for low quantity
of DNA) and that there is no post-PCR manipulation of samples (gel electrophoresis); this saves time and money, and
avoids contamination as post-PCR tubes are never opened in the laboratory.
Real-time quantitative PCR has enormous (largely untapped) potential to improve noninvasive studies by identify-
ing samples with enough nuclear DNA to avoid genotyping errors. The amount of DNA necessary to avoid genotyp-
ing errors (allelic dropout) has been estimated to be approximately 100–600 pg by theoretical and empirical studies
(e.g. Taberlet et al. 1996; Morin et al. 2001). RTQ-PCR could improve noninvasive studies by excluding extremely low
quality samples and identifying samples at risk of having genotyping errors.
A single RTQ-PCR can identify species in addition to quantifying amplifiable DNA (Berry & Sarre 2007). Species
identification is possible if species-specific primers are used or if the targeted PCR product has a different melting curve
(Berry & Sarre 2007). RTQ-PCR could replace species identification methods, which often involve mtDNA analysis and
that currently are the standard first step in many noninvasive studies (e.g. Swango et al. 2006).
The first paper using RTQ-PCR on noninvasive samples was Morin et al. (2001). Subsequently, the same RTQ-PCR
was used on ape faeces to identify factors (e.g. temperature) and sample preservation methods (ethanol and silica) that
improve PCR amplification. Several recent papers report successful RTQ-PCR of DNA from faeces and urine,
although most papers involve testing for cancer genes or disease pathogens in humans or livestock (e.g. Inglis & Kalis-
chuk 2004; Queipo-Ortuno et al. 2006; Itzkowitz et al. 2007). These recent papers are highly encouraging and suggest
that RTQ-PCR from faeces and urine is highly feasible and efficient.
We expect that RTQ-PCR will be widely used in future noninvasive studies because the methods have become eas-
ier (e.g. with commercial kits), less expensive, and clearly work on noninvasive samples (Hausknecht et al. 2007). An
RTQ-PCR reaction can cost as little as approximately US$1 per PCR (e.g. Berry & Sarre 2007). The cheapest RTQ-PCR
method (SYBR green) is also often highly reliable (e.g. Smith et al. 2002). An RTQ-PCR machine costs approximately
US$15 000–30 000 and prices are likely continue to fall (e.g. see http://www.biocompare.com/matrix/2838/Real-
Time-PCR-ermalCyclers(Thermocyclers).html).
� 2009 Blackwell Publishing Ltd
1288 T E C H N I C A L R E V I E W
an initial large-volume PCR with all primer pairs is per-
formed followed by a second or nested PCR of each
genetic marker (Piggott et al. 2004). The use of this two-
step PCR approach revealed significant improvements in
efficiency relative to standard PCR (Piggott et al. 2004;
Hedmark & Ellegren 2005; Arandjelovic et al. in press).
Because it requires less DNA extract, multiplex pre-
amplification allows typing more loci, which is often a
limitation in noninvasive genetics.
However, multiplex pre-amplification has drawbacks.
Allelic dropout can occur more frequently than for con-
ventional PCR, suggesting that this type of error is often
generated during the first-step multiplex (Lampa et al.
2008). In addition, a multiplex might increase the propor-
tion of nonamplifiable loci because of the competition
between loci (Lampa et al. 2008). Alternatively, genotypes
can be obtained by performing additional single standard
PCR whenever single locus amplification remains the
most suitable approach to satisfy efficiency and accuracy
(Parsons 2001). Although nested PCR increases the effi-
cacy and sensitivity for amplifying target genomic frag-
ments, it has the drawback of increasing the risk of
contamination, because it requires two PCR reactions
and, consequently, doubles the handling of materials.
This problem might be particularly prominent for nonin-
vasive studies.
Overcoming nonspecific amplification andcontamination
Co-amplification of nonspecific products and contamina-
tion can be major problems in noninvasive genetics. PCR
with low quality and quantity target DNA can increase
the probability of amplifying nontarget regions. It also
increases the probability that contaminant DNA is at simi-
lar or higher concentrations than target DNA (Pompanon
et al. 2005). Navidi et al. (1992) estimated that sporadic
contamination could cause up to 7% error in large-scale
studies, and Buchan et al. (2005) estimated that 1.3% of the
baboon DNA analysed and 1.2% of the negative controls
of their study were contaminated with human DNA.
Hot start PCR is one of the most effective means to
improve specificity, fidelity and sensitivity in DNA
amplifications. Effective protocols are now widely avail-
able thanks to the use of engineered thermostable
polymerases (whether using an inhibitor antibody or
chemical modification) that require heat activation prior
to PCR cycling, and because of the use of high-perfor-
mance PCR buffers with optimized combinations of salts
and additives (e.g. Radstrom et al. 2008). Taq polymerases
such as AmpliTaq GoldTM (Applied Byosistems), Fast-
Start Taq DNA Polymerase (Roche), Platinum� Taq DNA
polymerase (Invitrogen), TrueStart� Taq DNA Polymer-
ase (Fermentas), AccuSure� DNA Polymerase (Bioline),
Phusion� High-Fidelity DNA Polymerase (Finnzymes)
are a list of good examples (see Box 2).
DNA is present everywhere in a laboratory, especially
where PCRs are frequently performed because amplified
fragments persist as aerosols. Design of species-specific
primers reduces the risk of amplification of nonspecific
fragments and external DNA from human, prey items or
bacteria (particularly in faecal material). Primers that do
not amplify nontarget species (e.g. humans) can be
designed. This is increasingly feasible thanks to increas-
ing availability of sequence data from many species and
software programs to align and compare multiple
sequences.
Improved primer design with conventional software,
such as Primer 3 and a number of later adaptations
(Rozen & Skaletsky 2000; Kim & Lee 2007; Koressaar &
Remm 2007), PERLPRIMER (Marshall 2007) or SNPBOX (Weckx
et al. 2005) and highly specific multiplex primer design
tools are now available on the web. The server Primersta-
tion for the human genome (http://ps.cb.k.u-tokyo.ac.jp
Yamada et al. 2006), the program MULTIPLX (Kaplinski
et al. 2005) and the packages PRIMO (from BioToolKit 320;
Chang Bioscience) and PrimerPremier (PREMIER Bio-
soft) are examples of effective ways for designing specific
primers in large-scale analyses.
Precautions such as those in ancient DNA laboratories
should be followed to prevent and monitor for contami-
nation. Gilbert et al. (2005) describe nine criteria for work-
ing with ancient DNA and categorize risk factors
associated with different projects. The criteria include iso-
lation of work areas, use of negative controls for extrac-
tions and amplifications, amplification of only small
segments, reproducibility, use of cloning of products to
assess damage and contamination, independent replica-
tion, preservation of co-occurring biomolecules, quantifi-
cation of DNA and evaluation of associated remains.
They also consider hominid projects being the riskiest,
followed by projects on cultivars and domestic animals,
with low-risk projects involving projects on other wildlife
species. Among the most important precautions, PCR set
up should never be performed in the same day or just
after conducting PCR or entering a room with PCR
machines or post-PCR samples (see Fig. 1). Amplifying
additional loci that work in possible contaminant species
might also allow identifying contamination that remains
undetected in the analysis of the target markers (e.g.
Wandeler et al. 2003). For example, because of the high
copy number of mitochondrial molecules, using mito-
chondrial specific primers in both samples and controls
may be a sensitive way to monitor for contamination
when working with nuclear DNA (Pusch et al. 1998).
Design of PCR protocols that minimize manipulation
can reduce contamination risk. One could, for example,
develop multilocus assays to successfully work using the
� 2009 Blackwell Publishing Ltd
T E C H N I C A L R E V I E W 1289
minimum number of single-tube reactions, as it would
imply less manipulation for higher quantity of data pro-
duced per sample. RTQ-PCR has no post-PCR handling
(e.g. gel electrophoresis) and so tubes are not opened after
PCR, which minimizes DNA molecules in the laboratory
(Nazarenko et al. 2002). RTQ-PCR also allows the real-
time monitoring of target DNA amplification (Box 1) as
well as direct scoring of the desired results (e.g. melt curve
analysis, which can detect nontarget amplification).
Negative controls are essential to monitor contamina-
tion. Several blanks should be placed in the beginning (to
monitor for environmental and ⁄ or reagents contamina-
tions) and in the middle and end (to detect cross-contam-
ination) of a series of samples (Borst et al. 2004).
Minimizing PCR cycles (e.g. to 35 cycles) can reduce con-
tamination risks because tiny amounts of contamination
would unlikely lead to visible PCR products on electro-
pherograms or gels. Human forensic laboratories typi-
cally limit their PCR cycles to <35. However, this can be
problematic for degraded DNA samples, which can
require 40–45 PCR cycles.
Mixed samples can cause errors in noninvasive genet-
ics but can be detected and avoided using recent tech-
niques and software (Roon et al. 2005). Great efforts are
made to solve problems of DNA mixtures because more
than one donor is frequently responsible for the material
recovered from a forensic scene (e.g. in a rape, DNA from
the victim and the aggressor might be collected simulta-
neously). In this context, novel computational programs
have been developed to separate admixed genotypes,
such as PENDULUM (Bill et al. 2005) or MAIES (Cowell et al.
2006) that are based on different models to analyse peak
area values on electropherograms. DNA mixture should
not be regarded as a major limitation, because, if >6–8
highly polymorphic microsatellites are genotyped, it is
likely that some loci will have three alleles, which is
impossible for diploid species, and thus would indicate
possible contamination. Many wildlife and conservation
based studies that identify mixed samples simply discard
these samples in favour of those that indicate only one
animal deposited the sample.
Post-PCR and genotyping errors
The most insidious problem in noninvasive genetics is
genotyping errors. We define a genotyping error as a dif-
Box 2. Polymerase enzymes for PCR
Presently, there are several hundred companies selling over 20 kinds of polymerase enzymes. There are two main
characteristics that a polymerase enzyme must have that are crucial for amplifying small amounts of DNA: fidelity
and 3¢ fi 5¢ exonuclease activity (proofreading). Fidelity is particularly important when sequencing to detect SNP’s.
Heterozygous nucleotide sites must be unambiguously identified (in diploid individuals) or, for example, the false
discovery rate of SNPs might be high.
Proofreading with 3¢ fi 5¢ exonuclease activity is lacking in some polymerases [e.g. in Thermus aquaticus (Taq)] and
sequencing error rates are higher than for polymerases with exonuclease activity [e.g. isolated from Pyrococcus furiosus
(Pfu), Thermococcus litoralis (Vent), Pyrococcus woesei (Pwo)], which are often designated as high-fidelity polymerases.
Studies comparing regular Taq polymerase vs. high-fidelity polymerases, such as the Pfu, report far lower error rates
for the high-fidelity enzymes (Hansen et al. 2001).
Microsatellite genotyping with high-fidelity polymerases also gives lower error rates (Hite et al. 1996). When geno-
typing microsatellite loci (mostly dinucleotide), annoying stutter products are often formed during the PCR amplifica-
tion. The primary cause of ‘stutter’ bands is a change in the number of repeat units because of slip-strand extension by
Taq DNA polymerase. However, the use of high-fidelity polymerases (e.g. Pfu, Vent) reduces the formation of stutters
as 3¢ fi 5¢ exonuclease activity removes 3¢ nontemplate nucleotides (Hite et al. 1996).
A study testing different polymerase enzymes (Spitaleri et al. 2004), showed that, for low template quantities, the
regular Taq polymerases perform poorly and, for example, can increase allele dropout rates. However, in the same
study, the engineered polymerases (e.g. AmpliTaq Gold) maintained high fidelity and sensitivity at very low DNA
concentrations.
Amplification performance is another important characteristic. In this respect, it is well demonstrated that engi-
neered DNA polymerases perform much better with low quality DNA. This is mainly because engineered DNA
polymerases allow for the PCR hot-start technique. Hot start greatly increases the specificity and sensitivity of DNA
amplification by avoiding competing side reactions during pre-PCR setup that can be initiated the moment that all
reactants have been mixed and misprimming occurs.
At least two kinds of inactive polymerases are presently commercialized and often used in noninvasive studies: (i)
recombinant DNA polymerase (e.g. AmpliTaq Gold� Taq DNA Polymerase; Roche Molecular Systems) engineered to
be activated at temperatures higher than 90 �C, and (ii) Anti-Taq DNA polymerase antibodies, which inhibit polymer-
ase activity at room temperature (e.g. Platinum� Taq DNA Polymerase; Invitrogen).
� 2009 Blackwell Publishing Ltd
1290 T E C H N I C A L R E V I E W
ference between the true genotype and the inferred geno-
type (Pompanon et al. 2005; Luikart et al. 2008b), which
does not include failed PCRs or failed DNA extractions.
Amplification failure (no PCR product) is not as prob-
lematic as a genotyping error (erroneous genotype)
because mistakes in data interpretation are less likely
from failed PCRs. Genotyping error detection and avoid-
ance (e.g. by using the multi-tubes approach) have been
thoroughly reviewed elsewhere (Pompanon et al. 2005),
and so below we summarize and update the available
information, and highlight the main problems and ways
to avoid them.
Three main kinds of genotyping errors are generally
reported as: (i) allelic dropout (stochastic detection of
false homozygotes at heterozygous loci because of failure
of one allele to amplify), (ii) false alleles (creation of new
alleles caused by slippage events of Taq polymerase
during early cycles of PCR, that may reach a concentra-
tion similar to the authentic alleles when limited template
exists), and (iii) Human error, the incorrect identification
of alleles as a result of cross-contamination in the field or
in the laboratory or database manipulation errors (Hoff-
man & Amos 2005; Pompanon et al. 2005). Human errors
in data entry and manipulation (e.g. in spread sheets) are
often the most frequent cause of genotyping errors (Pae-
tkau 2003; Schwartz et al. 2006). Among the nonhuman-
induced errors, allelic dropout is usually the most
common error.
Extremely dissimilar error rates (depending on spe-
cies, season of the year and sample type) have been docu-
mented, ranging from as low as 0–2% in faecal analysis
(Bonin et al. 2004; Maudet et al. 2004) and 10% in human
buccal samples (Whitaker et al. 2001) to approximately
24% in some carnivore faeces (Johnson & Haydon 2007),
and over 30% in shed hairs (Gagneux et al. 1997). How-
ever, comparison of rates is challenging as some laborato-
ries are more conservative in discarding samples, while
others readily discard samples that show even the slight-
est sign of failure. These decisions dramatically change
the reported error rate.
There are four main approaches used to handle
genetic errors from noninvasive samples. The first and
the most common is called the multiple tubes approach
first developed by Navidi et al. (1992) and Taberlet et al.
(1996), which suggests that 6–10 similar genotypes
should be obtained for a locus to define an individual as
homozygous or heterozygous (see also Miller et al. 2002).
Here, each sample at each locus is run multiple times to
ensure genotype consistency. Some form of this approach
is used in almost every noninvasive study. However,
while multi-tubing will detect genotyping errors, it can
exhaust the DNA extracted and is fiscally expensive. In
addition, the multi-tube approach may increase errors as
samples are handled more often (inducing human error)
and there are more chances to produce false alleles,
which can be interpreted as a missing allele (allelic drop-
out). In addition, multi-tubing does nothing to prove that
the existing database is error free.
A second approach is to quantify the amount of target,
amplifiable nuclear DNA in the sample (Morin et al.
2001). Once this quantity is known, the appropriate num-
ber of multi-tube re-runs can be conducted. Morin et al.
(2001) recommended that if a sample has <25 pg (of ampl-
ifiable DNA) per reaction, it should be discarded; if it has
26–100 pg per reaction, then seven repeat genotypings of
the sample are necessary; if it has 101–200 pg per reaction
then four repeats are required; and if >200 pg per reac-
tion, only two repeats are necessary (see also Box 1).
A third approach has been to use computer algo-
rithms to detect genotyping errors. Depending on the
data and goal of the study, various algorithms have been
suggested (Ewen et al. 2000; Miller et al. 2002; Valiere
2002; Van Oosterhout et al. 2004; McKelvey & Schwartz
2005; Kalinowski 2006). Some of these examine devia-
tions from Hardy–Weinberg proportions, others use ped-
igree information to catch errors, while others use the
number of mismatches in recaptures (i.e. genotypes iden-
tified more than once and differing by only one or two
alleles; McKelvey & Schwartz 2005) as an error signal. A
recent paper suggests that sample-specific errors (only a
few poor quality individual samples) can cause signifi-
cant deviations from Hardy–Weinberg proportions; such
samples should be identified and often discarded
(Miquel et al. 2006). Some of the most widely used soft-
ware tools for detecting and avoiding genotyping errors
are provided in Table 2.
The fourth error handling approach is to model vari-
ous error rates in the final statistical analysis. For exam-
ple in capture–mark–recapture studies, Lukacs &
Burnham (2005) derived a method to incorporate the
probability of genotyping error into the closed-popula-
tion models of Otis et al. (1978), Huggins (1989) and Pled-
ger (2000) using the disproportionate number of
genotypes collected once relative to genotypes collected
more frequently to estimate error. These approaches have
been developed for estimating animal abundance, but are
relatively rare in population genetic studies. Another
example is in parentage studies where accommodating
genotyping errors during likelihood computations can
improve paternity analyses, as has been shown using the
software Cervus (Kalinowski et al. 2006). In a related
study, Wang (2004) developed likelihood methods to
infer full- and half-sibships from marker data with a high
error rate and to identify typing errors at each locus in
each reconstructed sib family.
It is important to note that blood and tissue samples
are too often assumed to always yield low genotyping
error rates. However, error rates can be substantial if
� 2009 Blackwell Publishing Ltd
T E C H N I C A L R E V I E W 1291
these (normally high quality) samples are poorly pre-
served (Hoffman & Amos 2005). Comparative analysis of
genotyping errors for noninvasive and assumed good
quality DNA is helpful and needed (Soulsbury et al.
2007), but should be interpreted with caution.
Regarding this, Johnson & Haydon (2007) developed a
maximum-likelihood-based method for estimating error
rates from a single replication of a sample of genotypes.
Simulations show it to be accurate and robust. It is imple-
mented in a computer program, PENDANT, which esti-
mates allelic dropout and false allele error rates with 95%
confidence regions from microsatellite genotype data and
performs power analysis. Finally, as mentioned in the
previous section, mixed samples (with DNA from more
than one individual) can be identified and computational
programs have been developed to resolve genotypes,
such as PENDULUM (Bill et al. 2005) or MAIES (Cowell et al.
2006).
Perspectives
The most promising areas for future research and devel-
opment in noninvasive genetic studies involve large-
scale PCR multiplexing techniques, massively parallel
sequencing technologies, and more holistic studies
including diet and parasite or disease analyses. Future
multiplexing techniques should allow analysis of tens to
hundreds of loci (Porreca et al. 2007; Meyer et al. 2008)
on noninvasive samples (see also Box 3). This would
vastly increase the statistical power of noninvasive
approaches and facilitate use of massively parallel
sequencing while making possible the targeted sequenc-
ing of interesting segments of the genome (e.g. exons
under selection).
New SNP multiplex genotyping systems use tiny
volumes (nanolitres) for SNP genotyping assays (e.g.
TaqMan; ABI), which reduces the costs of reaction chemi-
cals by nearly 98%, while automating and speeding up
the genotyping process. For example, a new multiplex
system using SNP chips from Fluidigm at BioMark�(http://www.fluidigm.com/applications/ genotype-pro
filing.html) allows simultaneous genotyping of 48 or 96
SNP loci on each of 48 or 96 individuals at a cost of only
US$0.10–0.20 per SNP (Perkel 2008). These systems,
however, also require an initial investment in equipment
often of the order of US$50 000–300 000.
Massively parallel sequencing technologies, e.g. 454
pyrosequencing by synthesis, and sequencing by ligation
(Ellegren 2008; Shendure & Li 2008), should improve
noninvasive studies because they work well on short
DNA fragments typical of difficult and ancient DNA
(Green et al. 2006). The main disadvantage of these
sequencing technologies is that they do not allow easy
sequencing of many individuals (samples), and the cost
per sequencing run is thousands of dollars. However,
costs are declining and clever study design can allow an
entire study to be conducted on a single sequencing run,
thereby minimizing total costs.
Table 2 Some examples of the most widely used methods and software programs developed mainly for detecting and preventing
genotyping errors
Software
Main functions
References
Identifying
problematic
samples
Estimating the
number of
multitube repeats
Identifying problematic
loci allelic dropout,
null alleles
Testing for
HWE departures
Identifying
mixed
samples
Quality Indexes* 4 4 Miquel et al. (2006)
Gemini† 4 4 4 4 Valiere et al. (2002)
Hw-QuickCheck 4 Kalinowski (2006)
Pedmanager‡ 4 4 4 Ewen et al. (2000)
Cervus 4 4 4 Marshall et al. (1998)
Gimlet 4 4 Valiere (2002)
Reliotype 4 4 4 Miller et al. (2002)
MICRO-CHECKER 4 4 4 Van Oosterhout et al. (2004)
Dropout§ 4 4 McKelvey & Schwartz (2005)
PENDULUM 4 Bill et al. (2005)
Pedant 4 Johnson & Haydon (2007)
GENECAP¶ 4 4 4 4 Wilberg & Dreher (2004)
*Program available upon request from the authors.
†Simulation-based method to detect consensus genotypes.
‡When pedigree information is available.
§Bimodal test for loci that cause many samples to differ by only one allele.
¶This is just a Microsoft Excel macro.
� 2009 Blackwell Publishing Ltd
1292 T E C H N I C A L R E V I E W
Increasingly holistic noninvasive genetic studies are
possible. They combine multiple kinds of information
(e.g. on diet, parasite load, parasite population genetics,
as well as host genetics and physiological status [stress
and reproductive hormone secretions]) allowing more
valuable studies addressing multiple questions or by
providing more complete information on individuals
allowing new questions to be addressed. Valentini et al.
(2009) used 454 pyrosequencing on chloroplast DNA
from faecal samples to determine the diet of bears (as
Box 3. Multiplex PCR techniques
Multiplex PCR amplification has great untapped potential to improve noninvasive sampling by reducing cost, increas-
ing speed and reducing consumption of DNA from typically low quantity sources (Henegariu et al. 1997; Butler 2005).
Reducing manipulation and handling (fewer PCRs per individual sample) also minimizes the possibility of contami-
nation and error during reaction setup.
Optimization of multiplex assays generally requires more time and effort than standard single plexes, because it
involves designing primer pairs that do not interact and at the same time anneal under the same conditions. Optimiza-
tion also sometimes requires, adjusting primer pair concentrations to give similar amounts of PCR product, choosing
fluorescence labels for sets of loci according to their allele or size range, and combining all these aspects in an efficient
and low-cost protocol (e.g. Guo & Milewicz 2007). Whenever possible, loci more difficult to amplify should be labelled
with the highest energetic labels (e.g. blue fluoresces brighter than red). Once obtained, multiplexes greatly facilitate
genotyping of large population samples rapidly and at reduced cost.
In forensics and noninvasive genetic studies, multiplex PCR is being used more for both microsatellites and SNPs
(Morin & Mccarthy 2007). Rapid and economical multiplex assays also exist for monitoring the international trade of
protected species; for example, a multiplex of several species-specific primers allows the distinction among shark spe-
cies (Shivji et al. 2005; Magnussen et al. 2007). Multiplexes have also been designed to study natural animal popula-
tions, e.g. a multiplex of 14 microsatellites in one PCR was developed for racoon, Procyon lotor, Fike et al. (2007).
Three main issues can facilitate multiplex PCR on noninvasive samples: (i) Recently developed commercial kits can
facilitate co-amplification of 5–10 loci or more (Luikart et al. 2008b). These kits include a new buffer that reduces com-
petition among loci and improves primer annealing. Multiplex PCR can be >30% cheaper than standard singleplex
(Mukherjee et al. 2007); (ii) The use of algorithms and software to design improved primer sets with no primer interac-
tions (Kaderali et al. 2003; Vallone & Butler 2004); and (iii) The use of universal fluorescent tails on the 5¢ end of prim-
ers to label PCR products (Oetting et al. 1995; Neilan et al. 1997). Fluorescent labelling of one primer in a pair is
expensive, ranging between US$100 and 150 (Schuelke 2000).
To reduce costs, Oetting et al. (1995) developed a single reaction nested PCR that allows easy and consistent geno-
typing and more homogeneous PCR amplification among loci. For each locus, PCR includes three different primers:
a reverse primer, a forward primer with a 5¢ tail (e.g. M-13 sequence), and the universal M-13 primer with fluores-
cent-labelling. During the first PCR cycles, the forward primer with tail hybridizes with the target DNA fragments
and is incorporated into the products, and then temperature is lowered (53 �C) to allow the universal tail to anneal
and incorporate fluorescence to the subsequent PCR products. With this technique, one can synthesize and use one
labelled forward primer (M-13) for each of several loci in a multiplex PCR (Missiaggia & Grattapaglia 2006). At the
same time, PCR multiplex amplification will be facilitated as the same forward primer (M-13) can give more even
amplification among loci and provide better results for low template DNA (Schuelke 2000). Laboratories studying
many species can benefit a lot from using a common universal labelled tail or tails. A cost reduction of �40% can be
achieved in the amplification of 10 microsatellites when compared with conventional methods (Missiaggia & Gratt-
apaglia 2006).
Most studies use the M13 sequence as the universal tail, but any sequence with no complementarity to target gen-
ome could be used (Neilan et al. 1997). For multiplexing several loci where some of them have overlapping size
ranges, one can optimize the PCR reaction using different fluorescent tails (Missiaggia & Grattapaglia 2006; Guo &
Milewicz 2007).
Single nucleotide polymorphism multiplex assays can work well using low quantity DNA, for example, 50 pg
(Onofri et al. 2006). Mini-STRs (up to 150 bp) have also been penta-plexed revealing detection limits of 12.5 pg for arti-
ficially degraded human DNA (Meissner et al. 2007). In noninvasive wildlife studies, multiplex PCR is not widely
used. However, Mukherjee et al. (2007) developed a multiplex protocol to identify tiger species from faeces using three
small mtDNA fragments. The multiplex had a significant decrease in the number of false negatives compared with
conventional PCR (especially in old faeces).
� 2009 Blackwell Publishing Ltd
T E C H N I C A L R E V I E W 1293
well as birds, snails and grasshoppers). They showed that
DNA-based faecal diet analysis using universal primers
(e.g. DNA barcoding) and pyrosequencing can help
determine what plant species are consumed by an indi-
vidual. In the future, noninvasive collection of spatially
referenced faeces from across a landscape could allow a
comprehensive study of a species (e.g. bears) in an area
by the enumeration of individuals, identification of gen-
der, examination of diet, estimation of parasite load and
parasite transmission patterns among individuals and
geographic areas. This type of information could become
crucial to the management of species and their habitat.
Concluding remarks
Application of noninvasive genetic approaches is exciting
and promising. The power and role of noninvasive genet-
ics in molecular ecology and conservation genetics will
continually increase, thanks to the advancements in each
step of a noninvasive study (Fig. 1) including new tech-
nologies (e.g. massively parallel sequencing) and
advancements from different disciplines (e.g. human and
livestock health, and forensics). Nonetheless, noninvasive
genetic studies still usually require more funding and
efforts in the laboratory, compared with traditional
genetic studies with high-quality DNA, to ensure low
genotyping error rates. Monitoring the efficacy and error
rate associated with each of the multiple steps in a nonin-
vasive study is crucial to ensure success.
Among the greatest needs for additional research is to
directly compare the relative performance of new and
improved methods (e.g. for sample storage, DNA extrac-
tion and amplification) in multiple independent laborato-
ries, taxa and sample types. The lack of independent and
quantitative comparisons of techniques makes it difficult
to provide advice on which methods are best for a given
species, sample type and sample conditions (but see Sch-
wartz & Monfort 2008, p. 242). Some techniques might be
species-specific and environment dependent, but more
studies are needed to assess this issue.
Research questions, including those that could be
addressed previously only using high-quality samples,
can now be addressed using noninvasive genetics, thanks
to lower error rates and our ability to analyse more loci
and more samples. For example, in many natural popula-
tions, it is increasingly feasible to estimate relatedness,
infer parentage and reconstruct pedigrees, all of which
require many loci and low genotyping error rates.
Genetic monitoring (Schwartz et al. 2007), defined as the
quantification of temporal changes in DNA-based esti-
mators (e.g. population abundance or effective size), is
also becoming more feasible because more samples can
be genotyped with more loci, thereby increasing statisti-
cal power to detect reduced variation, changes in popula-
tion size and immigration. In addition, noninvasive
genetics continually improves the ability of law enforce-
ment to detect illegal trafficking of animals (e.g. Manel
et al. 2002) by providing more representative samples
across populations and increasing recovery of DNA from
confiscated samples.
We are on the cusp of answering long-standing eco-
logical and evolutionary questions in rare and elusive
species, thanks to improved noninvasive sampling and
new technologies for analysing short DNA fragments
(Morin & Mccarthy 2007; Millar et al. 2008). This
includes questions about the genetic basis of local
adaptation that can be addressed by using genome-
wide scans (Wiehe et al. 2006) and population genomic
approaches (Luikart et al., 2003) requiring genotyping
of many loci, which is becoming feasible in noninva-
sive genetics. It also includes questions about how
landscape features influence gene flow and dispersal
in natural populations, which is a main goal of land-
scape genetics, an emerging approach that combines
landscape ecology and population genetics (Manel
et al. 2003). Landscape genetics typically requires anal-
yses of hundreds of samples widely dispersed across
landscapes; this is feasible only via noninvasive genetic
approaches in some taxa.
In disease ecology, we will be able to estimate trans-
mission rates and address questions about landscape fea-
tures or environmental variables influencing disease
spread, by noninvasively sampling of parasites (or para-
site DNA) from hosts (Archie et al. 2009). For example,
many microparasites (bacteria and viruses) and macro-
parasites (helminthes) are environmentally transmitted
(shed into the environment) and can be obtained from
faeces, urine or saliva. We can even conduct population
genomic studies on parasites (e.g. to identify genes influ-
encing transmission or virulence) for wildlife disease that
are notoriously difficult to study because they require
capture of many individuals, which is difficult or impos-
sible, as described earlier for elusive, rare or dangerous
wildlife species.
Overall, the recent boom in technological advances is
rapidly advancing the relatively new field of noninvasive
genetics. These new technologies are often derived from
human-based fields such as medicine and genomics. The
challenge for molecular ecologists will be keeping up
with and integrating these rapidly changing fields and
technologies to aid in the study and monitoring of wild
populations.
Acknowledgements
This review was developed during weekly international journal
club using Skype (Skype.com) for discussions between research-
ers at the University of Montana in the USA and the University
� 2009 Blackwell Publishing Ltd
1294 T E C H N I C A L R E V I E W
of Porto in Portugal. We thank Kristy Pilgrim, Steve Amish and
three anonymous reviewers for constructive comments on this
manuscript. AB-P and RO were supported by FCT grants
SFRH ⁄ BPD ⁄ 17822 ⁄ 2004 and SFRH ⁄ BD ⁄ 24361 ⁄ 2005 respectively.
GL was supported by the Portuguese American Foundation for
Development, CIBIO, UP and FCT grants PTDC ⁄ CVT ⁄69438 ⁄ 2006 and PTDC ⁄ BIA-BDE ⁄ 64111 ⁄ 2006. MKS was sup-
ported by a Presidential Early Career Award for Science and
Engineering.
References
Adams J, Waits L (2007) An efficient method for screening faecal
DNA genotypes and detecting new individuals and hybrids in
the red wolf (Canis rufus) experimental population area.
Conservation Genetics, 8, 123–131.
Anzai-Kanto E, Hirata MH, Hirata RDC et al. (2005) DNA
extraction from human saliva deposited on skin and its use in
forensic identification procedures. Brazilian Oral Research, 19,
216–222.
Arandjelovic M, Guschanski K, Schubert G et al. Two-step multi-
plex polymerase chain reaction improves the speed and accu-
racy of genotyping using DNA from noninvasive and
museum samples. Molecular Ecology Resources, 9, 28–36.
Archie EA, Luikart G, Ezenwa V (2009) Infecting epidemiology
with genetics: a new frontier in disease ecology. Trends in Ecol-
ogy and Evolution, 24, 21–30.
Baker CS, Cooke JG, Lavery S et al. (2007) Estimating the number
of whales entering trade using DNA profiling and capture-
recapture analysis of market products. Molecular Ecology, 16,
2617–2626.
Balkiz O, Dano S, Barbraund C et al. (2007) Sexing Greater fla-
mingo chicks from feather bulb DNA. Waterbirds, 30, 450–453.
Ball MC, Pither R, Manseau M (2007) Characterization of target
nuclear DNA from faeces reduces technical issues associated
with the assumptions of low-quality and quantity template.
Conservation genetics, 8, 577–586.
Bellemain E, Taberlet P (2004) Improved noninvasive genotyp-
ing method: application to brown bear (Ursus arctos) faeces.
Molecular Ecology Notes, 4, 519–522.
Bellemain E, Swenson JE, Tallmon D, Brunberg S, Taberlet P
(2005) Estimating population size of elusive animals with
DNA from hunter-collected faeces: four methods for brown
bears. Conservation Biology, 19, 150–161.
Berry O, Sarre SD (2007) Gel-free species identification using
melt-curve analysis. Molecular Ecology Notes, 7, 1–4.
Bhagavatula J, Singh L (2006) Genotyping faecal samples of Ben-
gal tiger Panthera tigris tigris for population estimation: a pilot
study. BMC Genetics, 7, 48.
Bill M, Gill P, Curran J et al. (2005) PENDULUM—a guideline-based
approach to the interpretation of STR mixtures. Forensic Science
International, 148, 181–189.
Bjornerfeldt S, Vila C (2007) Evaluation of methods for single
hair DNA amplification. Conservation Genetics, 8, 977–981.
Blejwas KM, Williams CL, Shin GT, McCullough DR, Jaeger
MM (2006) Salivary DNA evidence convicts breeding male
coyotes of killing sheep. Journal of Wildlife Management, 70,
1087–1093.
Bonin A, Bellemain E, Eidensen PB et al. (2004) How to track and
assess genotyping errors in population genetic studies. Molec-
ular Ecology, 13, 3261–3273.
Boom R, Sol CJ, Salimans MM et al. (1990) Rapid and simple
method for purification of nucleic acids. Journal of Clinical
Microbiology, 28, 495–503.
Borst A, Box ATA, Fluit AC (2004) False-positive results and con-
tamination in nucleic acid amplification assays: suggestions
for a prevent and destroy strategy. European Journal of Clinical
Microbiology and Infection Diseases, 23, 289–299.
Bradley BJ, Doran-Sheehy DM, Vigilant L (2007) Potential for
female kin associations in wild western gorillas despite female
dispersal. Proceedings of the Royal Society of London B: Biological
Sciences, 274, 2179–2185.
Bremner-Harrison S, Harrison SWR, Cypher BL, Murdock JD,
Maldonado J (2006) Development of a single-sampling nonin-
vasive hair snare. Wildlife Society Bulletin, 34, 456–461.
Broquet T, Petit E (2004) Quantifying genotyping errors in non-
invasive population genetics. Molecular Ecology, 13, 3601–3608.
Broquet T, Berset-Braendli L, Emaresi G, Fumagalli L (2007a)
Buccal swabs allow efficient and reliable microsatellite geno-
typing in amphibians. Conservation Genetics, 8, 509–511.
Broquet T, Menard N, Petit E (2007b) Noninvasive population
genetics: a review of sample source, diet, fragment length and
microsatellite motif effects on amplification success and geno-
typing error rates. Conservation Genetics, 8, 249–260.
Buchan JC, Archie EA, Van Horn RC, Moss CJ, Alberts SC (2005)
Locus effect and sources of error in noninvasive genotyping.
Molecular Ecology Notes, 5, 680–683.
Bush K, Vinsky M, Aldridge C, Paszkowski C (2005) A compari-
son of sample types varying in invasiveness for use in DNA
sex determination in an endangered population of greater
Sage-Grouse (Centrocercus uropihasianus). Conservation Genetics,
6, 867–870.
Butler JM (2005) Constructing STR multiplex assays. Methods in
Molecular Biology, 297, 53–66.
Butler J, Shen Y, McCord B (2003) The development of reduced
size STR amplicons as tools for analysis of degraded DNA.
Journal of Forensic Sciences, 48, 1054–1064.
Campbell NR, Narum SR (2009) Quantitative PCR assessment of
microsatellite and SNP genotyping with variable quality DNA
extracts. Conservation Genetics, doi: 10.1007/s10592-008-9661-7.
Casper RM, Jarman SN, Deagle BE, Gales NJ, Hindell MA (2007)
Detecting prey from DNA in predator scats: a comparison
with morphological analysis, using Arctocephalus seals fed a
known diet. Journal of Experimental Marine Biology and Ecology,
347, 144–154.
Chakraborty A, Sakai M, Iwatsuki Y (2006) Museum fish speci-
mens and molecular taxonomy: a comparative study on DNA
extraction protocols and preservation techniques. Journal of
Applied Ichthyology, 22, 160–166.
Chu J-H, Wu H-Y, Yang J, Takenaka O, Lin Y-S (1999) Polymor-
phic microsatellite loci and low-invasive DNA sampling in
Macaca cyclopis. Primates, 40, 573–580.
Chung DT, Drabek J, Opel KL, Butler JM, McCord BR (2004) A
study on the effects of degradation and template concentration
on the amplification efficiency of the STR miniplex primer sets.
Journal of Forensic Sciences, 49, 733–740.
Cowell RG, Lauritzen SL, Mortera J (2006) MAIES: a tool for DNA
mixture analysis. 22nd Conference on Uncertainty in Artificial
Intelligence (UAI), pp. 90–97, Cambridge, MA, USA.
Creel S, Spong G, Sands JL et al. (2003) Population size estima-
tion in Yellowstone wolves with error-prone noninvasive mi-
crosatellite genotypes. Molecular Ecology, 12, 2003–2009.
� 2009 Blackwell Publishing Ltd
T E C H N I C A L R E V I E W 1295
Cross N, Pines MK, Rogers LJ (2004) Saliva sampling to assess
cortisol levels in unrestrained common marmosets and the
effect of behavioral stress. American Journal of Primatology, 62,
107–114.
Culjkovic B, Savic D, Stojkovic O, Romac S (2003) Poly(A) tailing
of ancient DNA: a method for reproducible microsatellite
genotyping. Analytical Biochemistry, 318, 124–131.
Deagle BE, Gales NJ, Evans K et al. (2007) Studying seabird diet
through genetic analysis of faeces: a case study on macaroni
penguins (Eudyptes chrysolophus). PLoS ONE, 2, e831.
Dreher BP, Winterstein SR, Scribner KT et al. (2007) Noninvasive
estimation of black bear abundance incorporating genotyping
errors and harvested bear. Journal of Wildlife Management, 71,
2684–2693.
Durnin ME, Palsbøll PJ, Ryfer OA, McCullough DR (2007) A reli-
able genetic technique for sex determination of giant panda
(Ailuropoda melanoleuca) from non-invasively collected hair
samples. Conservation Genetics, 8, 715–720.
Ellegren H (2008) Sequencing goes 454 and takes large-scale ge-
nomics into the wild. Molecular Ecology, 17, 1629–1631.
Ellison SR, English CA, Burns MJ, Keer JT (2006) Routes to
improving the reliability of low level DNA analysis using real-
time PCR. BMC-Biotechnology, 6, 33.
Elphinstone MS, Hinten GN, Anderson MJ, Nock CJ (2003) An
inexpensive and high-throughput procedure to extract and
purify total genomic DNA for population studies. Molecular
Ecology Notes, 3, 317–320.
Epps CW, Palsbøll PJ, Wehausen JD, Roderick GK, McCullough
DR (2006) Elevation and connectivity define genetic refugia
for mountain sheep as climate warms. Molecular Ecology, 15,
4295–4302.
Ewen KR, Bahlo M, Treloar SA et al. (2000) Identification and
analysis of error types in high-throughput genotyping. Ameri-
can Journal of Human Genetics, 67, 727–736.
Farrell LE, Roma J, Sunquist ME (2000) Dietary separation of
sympatric carnivores identified by molecular analysis of scats.
Molecular Ecology, 9, 1583–1590.
Feinstein J (2004) DNA sequence from butterfly frass and
exuviae. Conservation Genetics, 5, 103–104.
Ficetola GF, Miaud C, Pompanon F, Taberlet P (2008) Species
detection using environmental DNA from water samples. Biol-
ogy Letters, 4, 423–425.
Fike JA, Drauch AM, Beasley JC, Dharmarajan G, Rhodes
OE (2007) Development of 14 multiplexed microsatellite
loci for raccoons Procyon lotor. Molecular Ecology Notes, 7,
525–527.
Frantz AC, Pope LC, Carpenter P et al. (2003) Reliable microsat-
ellite genotyping of the Eurasian badger (Meles meles) using
faecal DNA. Molecular Ecology, 12, 1649–1661.
Gagneux P, Boesch C, Woodruff DS (1997) Microsatellite scoring
errors associated with noninvasive genotyping based on
nuclear DNA amplified from shed hair. Molecular Ecology, 6,
861–868.
Gilbert MT, Bandelt H-J, Hofreiter M, Barnes I (2005) Assessing
ancient DNA studies. Trends in Ecology & Evolution, 20, 541–
544.
Goossens B, Abdullah ZB, Sinyor JB (2004) Which nests to
choose: collecting shed hairs from wild orang-utans. Folia Pri-
matologica, 75, 23–26.
Goymann W (2005) Noninvasive monitoring of hormones in
bird droppings: physiological validation, sampling, extrac-
tion, sex differences, and the influence of diet on hormone
metabolite levels. Annals of the New York Academy of Sciences,
1046, 35–53.
Grant SFA, Steinlicht S, Nentwich U et al. (2002) SNP genotyping
on a genome-wide amplified DOP-PCR template. Nucleic Acids
Research, 30, e125.
Green RE, Krause J, Ptak SE et al. (2006) Analysis of one million
base pairs of Neanderthal DNA. Nature, 444, 330–336.
Green ML, Herzing DL, Baldwin JD (2007) Noninvasive method-
ology for the sampling and extraction of DNA from free-rang-
ing Atlantic spotted dolphins (Stenella frontalis). Molecular
Ecology Notes, 7, 1287–1292.
Guo D-C, Milewicz DM (2007) Pyrosequencing protocols. In:
Methods in Molecular Biology (eds Walker JM, Marsh S), pp. 57–
62. Humana Press, New York.
Hajkova P, Zemanova B, Bryja J et al. (2006) Factors affecting suc-
cess of PCR amplification of microsatellite loci from otter fae-
ces. Molecular Ecology Notes, 6, 559–562.
Handel CM, Pajot LM, Talbot SL, Sage GK (2006) Use of buccal
swabs for sampling DNA from nestling and adult birds. Wild-
life Society Bulletin, 34, 1094–1100.
Hansen AJ, Willerslev E, Wiuf C, Mourier T, Arctander P (2001)
Statistical evidence for miscoding lesions in ancient DNA tem-
plates. Molecular Biology and Evolution, 18, 262–265.
Hausknecht R, Gula R, Pirga B, Kuehn R (2007) Urine—a source
for noninvasive genetic monitoring in wildlife. Molecular Ecol-
ogy Notes, 7, 208–212.
Hayakawa S, Takenaka O (1999) Urine as another potential
source for template DNA in polymerase chain reaction (PCR).
American Journal of Primatology, 48, 299–304.
Hedmark E, Ellegren H (2005) A test of the multiplex pre-ampli-
fication approach in microsatellite genotyping of wolverine
faecal DNA. Conservation Genetics, 7, 289–293.
Hedmark E, Flagstad Ø, Segerstrom P et al. (2004) DNA-based
individual and sex identification from wolverine (Gulo gulo)
faeces and urine. Conservation Genetics, 5, 405–410.
Henegariu O, Heerema NA, Dlouhy SR, Vance GH, Vogt PH
(1997) Multiplex PCR: critical parameters and step-by-step
protocol. BioTechniques, 23, 504–511.
Hite JM, Eckert KA, Cheng KC (1996) Factors affecting
fidelity of DNA synthesis during PCR amplification of
d(C-A)n.d(G-T)n microsatellite repeats. Nucleic Acids
Research, 24, 2429–2434.
Hoffman D, Amos W (2005) Microsatellite genotyping errors:
detection approaches, common sources and consequences for
paternal exclusion. Molecular Ecology, 14, 599–612.
Hogan FE, Cooke R, Burridge CP, Norman JA (2008) Optimizing
the use of shed feathers for genetic analysis. Molecular Ecology
Resources, 8, 561–567.
Horvath MB, Martinez-Cruz B, Negro JJ, Kalmar L, Godoy JA
(2005) An overlooked DNA source for non-invasive genetic
analysis in birds. Journal of Avian Biology, 36, 84–88.
Hoss M, Kohn M, Paabo S, Knauer F, Schroder W (1992) Excre-
ment analysis by PCR. Nature, 359, 199.
Huggins RM (1989) On the statistical analysis of capture experi-
ments. Biometrika, 76, 133–140.
Immell D, Anthony RG (2008) Estimation of black bear abun-
dance using a discrete DNA sampling device. Journal of Wild-
life Management, 72, 324–330.
Inglis GD, Kalischuk LD (2004) Direct quantification of Campylo-
bacter jejuni and Campylobacter lanienae in faeces of cattle by
� 2009 Blackwell Publishing Ltd
1296 T E C H N I C A L R E V I E W
real-time quantitative PCR. Applied and Environmental Microbi-
ology, 70, 2296–2306.
Inoue E, Miho I-M, Osamu T, Toshisada N (2007) Wild
Chimpanzee infant urine and saliva sampled noninva-
sively usable for DNA analyses. Journal of Primatology, 48, 156–
159.
Itzkowitz SH, Jandorf L, Brand R et al. (2007) Improved fecal
DNA test for colorectal cancer screening. Clinical Gastroentero-
logy and Hepatology, 5, 111–117.
Iyengar A, Babu VN, Hedges S et al. (2005) Phylogeography,
genetic structure, and diversity in the dhole (Cuon alpinus).
Molecular Ecology, 14, 2281–2297.
Jeffery KJ, Abernethy KA, Tutin CEG, Bruford MW (2007) Bio-
logical and environmental degradation of gorilla hair and mi-
crosatellite amplification success. Biological Journal of the
Linnean Society, 91, 281–294.
Jensen T, Pernasetti FM, Durrant B (2003) Conditions for rapid
sex determination in 47 avian species by PCR of genomic
DNA from blood, shell-membrane blood vessels, and feathers.
Zoo Biology, 22, 561–571.
Johnson PCD, Haydon DT (2007) Maximum-likelihood estima-
tion of allelic dropout and false allele error rates from micro-
satellite genotypes in the absence of reference data. Genetics,
175, 827–842.
Jones R, Cable J, Bruford MW (2008) An evaluation of non-inva-
sive sampling for genetic analysis in northern European rep-
tiles. Herpetological Journal, 18, 32–39.
Kaderali L, Deshpande A, Nolan JP, White PS (2003) Primer-
design for multiplexed genotyping. Nucleic Acids Research, 31,
1796–1802.
Kalinowski ST (2006) HW-QuickCheck: a computer program for
checking genotypes for agreement with Hardy–Weinberg
expectations. Molecular Ecology Notes, 6, 974–979.
Kalinowski ST, Taper ML, Marshall TC (2006) Revising how the
computer program Cervus accommodates genotyping error
increases success in paternity assignment. Molecular Ecology,
16, 1099–1106.
Kaplinski L, Anderson R, Puurand T, Remm M (2005) MULTIPLX:
automatic grouping and evaluation of PCR primers. Bioinfor-
matics, 21, 1701–1702.
Kawai K, Shimizu M, Hughes RN, Takenaka S (2004) A non-inva-
sive technique for obtaining DNA from marine intertidal snail.
Journal of the Marine Biological Association of the UK, 84, 773–774.
Kendall KC, McKelvey KS (2008) Hair collection. In: Noninvasive
Survey Methods for North American Carnivores (eds Long RA,
MacKay P, Ray JC, Zielinski WJ), pp. 135–176. Island Press,
Washington, DC.
Kendall KC, Stetz JB, Boulanger J et al. (2009) Demography and
genetic structure of a recovering brown bear population. Jour-
nal of Wildlife Management, 73, 2–17.
Kerley LL, Salkina GP (2007) Using scent-matching dogs to iden-
tify individual Amur tigers from scats. Journal of Wildlife Man-
agement, 71, 1349–1356.
Kim N, Lee C (2007) QPRIMER: a quick web-based application for
designing conserved PCR primers from multigenome align-
ments. Bioinformatics, 23, 2331–2333.
Kittler R, Stoneking M, Kayser C (2002) A whole genome ampli-
fication method to generate long fragments from low quanti-
ties of genomic DNA. Analytical Biochemistry, 300, 237–244.
Kohn MH, Wayne RK (1997) Facts from faeces revisited. Trends
in Ecology and Evolution, 6, 223–227.
Kohn MH, York EC, Kamradt DA et al. (1999) Estimating popu-
lation size by genotyping faeces. Proceedings of the Royal Society
of London B: Biological Sciences, 266, 657–663.
Kontanis EJ, Reed FA (2006) Evaluation of real-time PCR ampli-
fication efficiencies to detect PCR inhibitors. Journal of Forensic
Sciences, 51, 795–804.
Koressaar T, Remm M (2007) Enhancements and modifications
of primer design program Primer 3. Bioinformatics, 23, 1289–
1291.
Koukoulas I, O’Toole FE, Stringer P, van Oorschot RAH (2008)
QuantifilerTM: observations of relevance to forensic casework.
Journal of Forensic Sciences, 53, 135–141.
Lampa S, Gruber B, Henle K, Hoehn M (2008) An optimisation
approach to increase DNA amplification success of otter fae-
ces. Conservation Genetics, 9, 201–210.
Lanyon CV, Rushton SP, O’Donnell AG et al. (2007) Murine scent
mark microbial communities are genetically determined.
FEMS Microbiology Ecology, 59, 576–583.
Lau LT, Fung YW, Wong FP et al. (2003) A real-time PCR for
SARS-coronavirus incorporating target gene pre-amplifica-
tion. Biochemical and Biophysical Research Communications, 312,
1290–1296.
Lecomte N, Gauthier G, Bernatchez L, Giroux J-F (2006) A
nondamaging blood sampling technique for waterfowl
embryos. Journal of Field Ornithology, 77, 67–70.
Lee PLM, Prys-Jones RP (2008) Extracting DNA from museum
bird eggs, and whole genome amplification of archive DNA.
Molecular Ecology Resources, 8, 551–560.
Leonard JA, Wayne RK, Cooper A (2000) Population genetics of
Ice Age brown bears. Proceedings of the National Academy of
Sciences of the United States of America, 97, 1651–1654.
Luikart G, England PR, Tallmon D, Jordan S, Taberlet P
(2003) The power and promise of population genomics:
from genotyping to genome typing. Nature Reviews in
Genetics, 4, 981–94.
Luikart G, Zundel S, Rioux D et al. (2008a) Low genotyping error
rates for microsatellite multiplexes and noninvasive fecal
DNA samples from bighorn sheep. Journal of Wildlife Manage-
ment, 72, 299–304.
Luikart G, Pilgrim K, Visty J, Ezenwa VO, Schwartz MK
(2008b) Candidate gene microsatellite variation is associated
with parasitism in wild bighorn sheep. Biology Letters, 4, 228–
231.
Lukacs PM, Burnham KP (2005) Review of capture-recapture
methods applicable to noninvasive genetic sampling. Molecu-
lar Ecology, 14, 3909–3919.
Magnussen JE, Pikitch EK, Clarke SC et al. (2007) Genetic track-
ing of basking shark products in international trade. Conserva-
tion Genetics, 10, 199–207.
Manel S, Berthier P, Luikart G (2002) Detecting wildlife poach-
ing: identifying the origin of individuals using Bayesian
assignment tests and multi-locus genotypes. Conservation Biol-
ogy, 16, 650–657.
Manel S, Schwartz MK, Luikart G, Taberlet P (2003) Landscape
genetics: combining landscape ecology and population genet-
ics. Trends in Ecology and Evolution, 18, 189–197.
Marshall O (2007) Graphical design of primers with PERLPRIMER.
Methods in Molecular Biology, 402, 403–414.
Marshall TC, Slate J, Kruuk LEB, Pemberton JM (1998) Statistical
confidence for likelihood-based paternity inference in natural
populations. Molecular Ecology, 7, 639–655.
� 2009 Blackwell Publishing Ltd
T E C H N I C A L R E V I E W 1297
Martinkova N, Searle JB (2006) Amplification success rate of
DNA from museum skin collections: a case study of stoats
from 18 museums. Molecular Ecology Notes, 6, 1014–
1017.
Maudet C, Luikart G, Dubray D, Von Hardenberg A, Taberlet P
(2004) Low genotyping error rates in wild ungulate faeces
sampled in winter. Molecular Ecology Notes, 4, 772–775.
Mayntz-Press KA, Ballantyne J (2007) Performance characteris-
tics of commercial Y-STR multiplex systems*. Journal of Foren-
sic Sciences, 52, 1025–1034.
McDaniel GW, McKelvey KS, Squires JR, Ruggiero LF (2000)
Efficacy of lures and hair snares to detect lynx. Wildlife Society
Bulletin, 28, 119–123.
McKay P, Smith DA, Long RA, Parker M (2008) Scat detection
dogs. In: Noninvasive Survey Methods for North American Carni-
vores (eds Long RA, MacKay P, Ray JC, Zielinski WJ), pp. 135–
176. Island Press, Washington, DC.
McKelvey KS, Schwartz MK (2005) Dropout: a program to iden-
tify problem loci and samples for noninvasive genetic samples
in a capture-mark-recapture framework. Molecular Ecology
Notes, 5, 716–718.
Meissner C, Bruse P, Mueller E, Oehmichen M (2007) A new sen-
sitive short pentaplex (ShoP) PCR for typing of degraded
DNA. Journal of Forensic Sciences, 166, 121–127.
Meyburg B-U, Meyburg C, Francj-Neumann F (2007) Why do
female Lesser Spotted Eagles (Aquila pomarina) visit strange
nests remote from their own? Journal of Ornithology, 148, 157–
166.
Meyer M, Stenzel U, Hofreiter M (2008) Parallel tagged sequenc-
ing on the 454 platform. Nature Protocols, 3, 267–278.
Millar CD, Huynen L, Subramanian S, Mohandesan E, Lambert
DM (2008) New developments in ancient genomics. Trends in
Ecology and Evolution, 23, 386–393.
Miller HC (2006) Cloacal and buccal swabs are a reliable source
of DNA for microsatellite genotyping of reptiles. Conservation
Genetics, 7, 1001–1003.
Miller C, Joyce P, Waits LP (2002) Assessing allelic dropout and
genotype reliability using maximum likelihood. Genetics, 160,
357–366.
Miller MM, Ealey KA, Oswald WB, Schat KA (2003) Detection of
chicken anemia virus DNA in embryonal tissues and eggshell
membranes. Avian Diseases, 47, 662–671.
Miquel C, Bellemain E, Poillot C, Taberlet P (2006) Quality
indexes to assess the reliability of genotypes in studies using
noninvasive sampling and multiple-tube approach. Molecular
Ecology Notes, 6, 985–988.
Missiaggia A, Grattapaglia D (2006) Plant microsatellite geno-
typing with 4-color fluorescent detection using multiple-tailed
primers. Genetics and Molecular Resources, 5, 72–78.
Mitrovski P, Heinze DA, Guthridge K, Weeks AR (2005) Isola-
tion and characterization of microsatellite loci from the Aus-
tralian endemic mountain pygmy-possum, Burramys parvus
Broom. Molecular Ecology Notes, 5, 395–397.
Mitrovski P, Heinze DA, Broome L, Hoffmann AA, Weeks AR
(2007) High levels of variation despite genetic fragmentation in
populations of the endangered mountain pygmy-possum, Bur-
ramys parvus, in alpine Australia. Molecular Ecology, 16, 75–87.
Morin P, Mccarthy M (2007) Highly accurate SNP genotyping
from historical and low quality samples. Molecular Ecology
Notes, 7, 937–946.
Morin PA, Chambers KE, Boesh C, Vigilant L (2001) Quantitative
polymerase chain reaction analysis of DNA from noninvasive
samples for accurate microsatellite genotyping of wild chim-
panzees (Pan troglodytes verus). Molecular Ecology, 10, 1835–
1844.
Morin PA, Luikart G, Wayne RK, The SNP Workshop Group
(2004) SNPs in ecology, evolution and conservation. Trends in
Ecology and Evolution, 19, 208–216.
Morin PA, Hedrick NM, Robertson KM, Leduc CA (2007)
Comparative mitochondrial and nuclear quantitative PCR
of historical marine mammal tissue, bone, baleen, and tooth
samples. Molecular Ecology Notes, 7, 404–411.
Morin PA, Martien KK, Taylor BL (2009a) Assessing statistical
power of SNPs for population structure and conservation
studies. Molecular Ecology Resources, 9, 66–73.
Morin PA, LeDuc RG, Archer FI et al. (2009b) Significant
deviations from Hardy-Weinberg equilibrium caused by low
levels of microsatellite genotyping errors. Molecular Ecology
Resources, 9, 498–504.
Morling N (2008) Handbook of forensic genetics. In: Forensic
Science and Medicine, p. 400. Humana Press, New York.
Mowat G, Paetkau D (2002) Estimating marten Martes americana
population size using hair capture and genetic tagging. Wild-
life Biology, 8, 201–209.
Mukherjee N, Mondol S, Andheria A, Ramakrishnan U (2007)
Rapid multiplex PCR based species identification of wild
tigers using non-invasive samples. Conservation Genetics, 8,
1465–1470.
Musgrave-Brown E, Ballard D, Balogh K et al. (2007) Forensic
validation of the SNPforID 52-plex assay. Forensic Science Inter-
national, 1, 186–190.
Navidi W, Arnheim N, Waterman MS (1992) A multiple-tubes
approach for accurate genotyping of very small DNA samples
by using PCR: statistical considerations. American Journal of
Human Genetics, 50, 347–359.
Nazarenko I, Lowe B, Darfler M et al. (2002) Multiplex quantita-
tive PCR using self-quenched primers labelled with a single
fluophore. Nucleic Acids Research, 30, e37.
Neilan BA, Wilton AN, Jacobs D (1997) A universal procedure
for primer labelling of amplicons. Nucleic Acids Research, 25,
2938–2939.
Nsubuga AM, Robbins MM, Roeder A, Morin P, Boesch C, Vigi-
lant L (2004) Factors affecting the amount of genomic DNA
extracted from ape feces and the identification of an improved
sample storage method. Molecular Ecology, 13, 2089–2094.
Oetting WS, Lee HK, Flanders DJ et al. (1995) Linkage analysis
with multiplexed short tandem repeat polymorphisms using
infrared fluorescence and M13 tailed primers. Genomics, 30,
450–458.
Onofri V, Alessandrini F, Turchi C et al. (2006) Development of
multiplex PCRs for evolutionary and forensic applications of
37 human Y chromosome SNPs. Forensic Science International,
157, 23–35.
Otis DL, Burnham KP, White GC, Anderson DR (1978) Statistical
inference from capture data on closed animal populations.
Wildlife Monographs, 62, 178.
Paetkau D (2003) An impirical exploration of data quality in
DNA-based population inventories. Molecular Ecology, 12,
181–184.
� 2009 Blackwell Publishing Ltd
1298 T E C H N I C A L R E V I E W
Palmer A, Styan C, Shearman D (2008) Foot mucus is a good
source for non-destructive genetic sampling in Polyplacopho-
ra. Conservation Genetics, 9, 229–231.
Palomares F, Godoy JA, Piriz A, O¢Brien SJ, Johnson WE (2002)
Faecal genetic analysis to determine the presence and distribu-
tion of elusive carnivores: design and feasibility for Iberian
lynx. Molecular Ecology, 11, 2171–2182.
Palsboll PJ, Allen J, Berub M et al. (1997) Genetic tagging of
humpback whales. Nature, 388, 767–769.
Pang BCM, Cheung BKK (2007) Double swab technique for
collecting touched evidence. Legal Medicine, 9, 181–184.
Parsons KM (2001) Reliable microsatellite genotyping of dol-
phin DNA from faeces. Molecular Ecology Notes, 1, 341–
344.
Pauli JN, Hamilton MB, Crain EB, Buskirk SW (2008) A single-
sampling hair trap for mesocarnivores. Journal of Wildlife Man-
agement, 72, 1650–1652.
Perkel J (2008) SNP genotyping: six technologies that keyed a
revolution. Nature Methods, 5, 447–453.
Pfeiffer I, Volkel I, Taubert H, Brenig B (2004) Forensic DNA-typ-
ing of dog hair: DNA-extraction and PCR amplification. Foren-
sic Science International, 141, 149–151.
Piggott MP, Taylor AC (2003) Extensive evaluation of faecal
preservation and DNA extraction methods in Australian
native and introduced species. Australian Journal of Zoology, 51,
341–355.
Piggott M, Bellemain EA, Taberlet P, Taylor AC (2004)
Multiplex pre-amplification method that significantly
improves microsatellite amplification and error rates for
faecal DNA in limiting conditions. Conservation Genetics, 5,
417–420.
Piggott MP, Banks SC, Stone N, Banffy C, Taylor AC (2006) Esti-
mating population size of endangered brush-tailed rockwal-
laby (Petrogale Penicillata) colonies using faecal DNA.
Molecular Ecology, 15, 81–91.
Pledger S (2000) Unified maximum likelihood estimates for
closed capture-recapture models using mixtures. Biometrics,
56, 434–442.
Pompanon F, Bonin A, Bellemain E, Taberlet P (2005) Genotyp-
ing errors: causes, consequences and solutions. Nature Reviews.
Genetics, 6, 847–846.
Porreca GJ, Zhang K, Li J et al. (2007) Multiplex amplification of
large sets of human exons. Nature Methods, 4, 931–936.
Puechmaille SJ, Mathy G, Petit EJ (2007) Good DNA from bat
droppings. Acta Chiropterologica, 9, 269–276.
Pusch CM, Giddings I, Scholz M (1998) Repair of degraded
duplex DNA from prehistoric samples using Escherichia coli
DNA polymerase I and T4 DNA ligase. Nucleic Acids Research,
26, 857–859.
Queipo-Ortuno MI, Colmenero JD, Munoz N et al. (2006) Rapid
diagnosis of Brucella epididymo-orchitis by real-time poly-
merase chain reaction assay in urine samples. The Journal of
Urology, 176, 2290–2293.
Radstrom P, Lofstrom C, Lovenklev M, Knutsson R, Wolffs P
(2008) Strategies for overcoming PCR inhibition. Cold Spring
Harbor Protocols 2008, Cold Spring Harbor Laboratory Press,
Cold Spring Harbor, NY.
Regnaut S, Lucas FS, Fumagalli L (2006) DNA degradation in
avian faecal samples and feasibility of non-invasive genetic
studies of threatened capercaillie populations. Conservation
Genetics, 7, 449–453.
Roeder AD, Archer FI, Poinar HN, Morin PA (2004) A novel
method for collection and preservation of faeces for genetic
studies. Molecular Ecology Notes, 4, 761–764.
Roon DA, Waits LP, Kendall KC (2003) A quantitative evaluation
of two methods for preserving hair samples. Molecular Ecology
Notes, 3, 163–166.
Roon DA, Thomas ME, Kendall K, Waits LP (2005) Evaluating
mixed samples as a source of error in noninvasive genetic
studies using microsatellites. Molecular Ecology, 14, 195–201.
Rozen S, Skaletsky HJ (2000) Primer 3 on the WWW for general
users and for biologist programmers. In: Bioinformatics Methods
and Protocols: Methods in Molecular Biology (eds Krawetz S &
Misener S), pp. 365–386. Humana Press, Totowa, NJ, USA.
Rudnick JA, Katzner TE, Bragin EA, DeWoody JA (2007) Species
identification of birds through genetic analysis of naturally
shed feathers. Molecular Ecology Notes, 7, 757–762.
Rudnick JA, Katzner TE, Bragin EA, DeWoody JA (2008) A non-
invasive genetic evaluation of population size, natal philopa-
try, and roosting behavior of non-breeding eastern imperial
eagles (Aquila heliaca) in central Asia. Conservation Genetics, 9,
667–676.
Ruiz-Gonzalez A, Rubines A, Berdion O, Gomez-Moliner BJ
(2008) A non-invasive genetic method to identify the sympat-
ric mustelids pine marten (Martes martes) and stone marten
(Martes foina): preliminary distribution survey on the northern
Iberian Peninsula. European Journal of Wildlife Research, 54, 253–
261.
Santini A, Lucchini V, Fabbri E, Randi E (2007) Ageing and envi-
ronmental factors affect PCR success in wolf (Canis lupus)
excremental DNA samples. Molecular Ecology Notes, 7, 955–
961.
Sastre N, Francino O, Lampreave G et al. (2009) Sex identification
of wolf (Canis lupus) using non-invasive samples. Conservation
Genetics, doi: 10.1007/s10592-008-9565-6.
Scandura M (2005) Individual sexing and genotyping from blood
spots on the snow: a reliable source of DNA for non-invasive
genetic surveys. Conservation Genetics, 6, 871–874.
Schmidt K, Kowalczyk R (2006) Using scent-marking stations to
collect hair samples to monitor Eurasian lynx populations.
Wildlife Society Bulletin, 34, 462–466.
Schuelke M (2000) An economic method for the fluorescent
labeling of PCR fragments. Nature Biotechnology, 18, 233–
234.
Schwartz MK, Monfort SL (2008) Genetic and endocrine tools for
carnivore surveys. In: Noninvasive Survey Methods for North
American Carnivores (eds Long RA, MacKay P, Ray JC, Zielin-
ski WJ), pp. 228–250. Island Press, Washington, DC.
Schwartz MK, Cushman SA, McKelvey KS, Hayden J, Engkjer C
(2006) Detecting genotyping errors and describing black bear
movement in North Idaho. Ursus, 17, 138–148.
Schmaltz G, Somers CM, Sharma P, Quinn JS (2006) Non-
destructive sampling of maternal DNA from the external shell
of bird eggs. Conservation Genetics, 7, 543–549.
Schwartz MK, Aubry KB, McKelvey KS et al. (2007) Inferring
geographic isolation of wolverine in California using historical
DNA. Journal of Wildlife Management, 71, 2170–2179.
Segelbacher G (2002) Noninvasive genetic analysis in birds: test-
ing reliability of feather samples. Molecular Ecology Notes, 2,
367–369.
Shendure J, Li H (2008) Next-generation DNA sequencing. Nat-
ure Biotechnology, 26, 1135–1145.
� 2009 Blackwell Publishing Ltd
T E C H N I C A L R E V I E W 1299
Shivji MS, Chapman DD, Pikitch EK, Raymond PW (2005)
Genetic profiling reveals illegal international trade in fins of
the great white shark, Carcharodon carcharias. Conservation
Genetics, 6, 1035–1039.
Smith S, Morin PA (2005) Optimal storage conditions for highly
dilute DNA samples: a role for trehalose as a preserving agent.
Journal of Forensic Sciences, 50, 1101–1108.
Smith S, Vigilant L, Morin PA (2002) The effects of sequence
length and oligonucleotide mismatches on 5¢ exonuclease
assay efficiency. Nucleic Acids Research, 30, e111.
Smith DA, Ralls K, Cypher BL, Maldonado JE (2005) Assessment
of scat-detection dog surveys to determine kit fox distribution.
Wildlife Society Bulletin, 33, 897–904.
Soulsbury C, Iossa G, Edwards K, Baker P, Harris S (2007) Allelic
dropout from a high-quality DNA source. Conservation Genet-
ics, 8, 733–738.
Spitaleri S, Piscitello D, Di Martino D (2004) Experimental proce-
dures comparing the activity of different Taq polymerases.
Forensic Science International, 146, S167–S169.
Sundqvist A-K, Ellegren H, Vila C (2008) Wolf or dog?
Genetic identification of predators from saliva collected
around bite wounds on prey. Conservation Genetics, 9, 1275–
1279.
Swango KL, Hudlow WR, Timken MD, Buoncristiani MR (2006)
A quantitative PCR assay for the assessment of DNA degrada-
tion in forensic samples. Forensic Science International, 158, 14–
26.
Swanson BJ, Kelly BP, Maddox CK, Moran JR (2006) Shed skin
as a source of DNA for genotyping seals. Molecular Ecology
Notes, 6, 1006–1009.
Taberlet P, Bouvet J (1992) Bear conservation genetics. Nature,
358, 197.
Taberlet P, Griffin S, Goossens B et al. (1996) Reliable genotyping
of samples with very low DNA quantities using PCR. Nucleic
Acids Research, 24, 3189–3194.
Taberlet P, Waits LP, Luikart G (1999) Noninvasive genetic sam-
pling: look before you leap. Trends in Ecology and Evolution, 14,
323–327.
Tang Y-W, Sefers SE, Li H, Kohn DJ, Procop GW (2005) Compar-
ative evaluation of three commercial systems for nucleic acid
extraction from urine specimens. Journal of Applied Microbiol-
ogy, 43, 4830–4833.
Thakuria D, Schmidt O, Liliensiek AK, Egan D, Doohan FM
(2009) Field preservation and DNA extraction methods for
intestinal microbial diversity analysis in earthworms. Journal
of Microbiological Methods, 76, 226–233.
Thomas M, Gilbert P (2006) Postmortem damage of mitochon-
drial DNA. In: Human Mitochondrial DNA and the Evolution of
Homo sapiens (eds Bandelt H-J, Macaulay V, Richards M), pp.
91–115. Springer-Verlag, Berlin.
Thornton CG, Passen S (2004) Inhibition of PCR amplification by
phytic acid, and treatment of bovine fecal specimens with phy-
tase to reduce inhibition. Journal of Microbiological Methods, 59,
43–52.
Toth M (2008) A new noninvasive method for detecting mam-
mals from birds nests. Journal of Wildlife Management, 72, 1237–
1240.
Ulizio T, Squires JR, Petscher DH et al. (2006) The efficacy
of obtaining genetic-based identifications from putative
wolverine snow tracks. Wildlife Society Bulletin, 34, 1326–
1332.
Valentini A, Miquel C, Nawaz MA et al. (2009) New perspectives
in diet analysis based on DNA barcoding and parallel pyrose-
quencing: the trnL approach. Molecular Ecology Resources, 9,
51–60.
Valiere N (2002) Gimlet: a computer program for analysing
genetic individual identification data. Molecular Ecology Notes,
2, 377–379.
Valiere N, Taberlet P (2000) Urine collected in the field as a
source of DNA for species and individual identification.
Molecular Ecology, 9, 2150–2152.
Valiere N, Berthier P, Mouchiroud D, Pontier D (2002) Gemini:
software for testing the effects of genotyping errors and multi-
tubes approach for individual identification. Molecular Ecology
Notes, 2, 83–86.
Valiere N, Bonenfant C, Toıgo C, Luikart G, Gaillard J-M, Klein F
(2006) Importance of a pilot study for non-invasive genetic
sampling: genotyping errors and population size estimation in
red deer. Conservation Genetics, 8, 69–78.
Vallet D, Petit EJ, Gatti S, Levrero F, Menard N (2008) A new
2CTAB ⁄ PCI method improves DNA amplification success
from faeces of Mediterranean (Barbary macaques) and tropi-
cal (lowland gorillas) primates. Conservation Genetics, 9, 677–
680.
Vallone PM, Butler JM (2004) AutoDimer: a screening tool for
primer-dimer and hairpin structures. BioTechniques, 37, 226–
231.
Van Oosterhout C, Hutchinson WF, Willis DPM, Shipley P
(2004) MICRO-CHECKER: software for identifying and correcting
genotyping errors in microsatellite data. Molecular Ecology
Notes, 4, 535–538.
Vigilant L (1999) An evaluation of techniques for the extraction
and amplification of DNA from naturally shed hairs. Biological
Chemistry, 380, 1329–1331.
Waits LP, Paetkau D (2005) Noninvasive genetic sampling tools
for wildlife biologists: a review of applications and recommen-
dations for accurate data collection. Journal of Wildlife Manage-
ment, 69, 1419–1433.
Walker FM, Sunnucks P, Taylor AC (2008) Evidence for habitat
fragmentation altering within-population processes in wom-
bats. Molecular Ecology, 17, 1674–1684.
Wan Q-H, Zhu L, Wu HUA, Fang S-G (2006) Major histocompat-
ibility complex class II variation in the giant panda (Ailuropoda
melanoleuca). Molecular Ecology, 15, 2441–2450.
Wandeler P, Smith S, Morin PA, Pettifor RA, Funk SM (2003)
Patterns of nuclear DNA degeneration over time—a case
study in historic teeth samples. Molecular Ecology, 12, 1087–
1093.
Wang J (2004) Sibship reconstruction from genetic data with typ-
ing errors. Genetics, 166, 1963–1979.
Wasser SK, Mailand C, Booth R et al. (2007) Using DNA to track
the origin of the largest ivory seizure since the 1989 trade ban.
Proceedings of the National Academy of Sciences of the United
States of America, 104, 4228–4233.
Weaver JL, Wood P, Paetkau D, Laack LL (2005) Use of scented
hair snares to detect ocelots. Wildlife Society Bulletin, 33, 1384–
1391.
Weckx S, De Rijk P, Van Broeckhoven C, Del-Favero J (2005) SNP-
BOX: a modular software package for large-scale primer design.
Bioinformatics, 21, 385–387.
Whitaker JP, Cotton EA, Gill P (2001) A comparison of the char-
acteristics of profiles produced with the AMPFlSTR(R) SGM
� 2009 Blackwell Publishing Ltd
1300 T E C H N I C A L R E V I E W
Plus(TM) multiplex system for both standard and low copy
number (LCN) STR DNA analysis. Forensic Science Interna-
tional, 123, 215–223.
Wiehe T, Nolte V, Zivkovic D et al. (2006) Non-destructive sam-
pling of maternal DNA from the external shell of bird eggs.
Conservation Genetics, 7, 543–549.
Wiehe T, Nolte V, Zivkovic D, Schlotterer C (2007) Identification
of selective sweeps using a dynamically adjusted number of
linked microsatellites. Genetics, 175, 207–218.
Wilberg MJ, Dreher BP (2004) GENECAP: a program for analysis of
multilocus genotype data for non-invasive sampling and cap-
ture-recapture population estimation. Molecular Ecology Notes,
4, 783–785.
Willard JM, Lee DA, Holland MM (1998) Recovery of DNA
for PCR amplification from blood and forensic samples
using a chelating resin. In: Forensic DNA Profiling Protocols
(eds Lincoln PJ, Thomson J), pp. 9–18. Humana Press,
Totowa, NJ.
Williams CL, Blejwas K, Johnston JJ, Jaeger MM (2003) A coyote
in sheep¢s clothing: predator identification from saliva. Wildlife
Society Bulletin, 31, 926–932.
Yamada T, Soma H, Morishita S (2006) PrimerStation: a
highly specific multiplex genomic PCR primer design ser-
ver for the human genome. Nucleic Acids Research, 34,
W665–W669.
Zielinski WJ, Schlexer FV, Pilgrim KL, Schwartz MK (2006) The
efficacy of wire and glue hair snares in identifying mesocarni-
vores. Wildlife Society Bulletin, 34, 1152–1161.
� 2009 Blackwell Publishing Ltd
T E C H N I C A L R E V I E W 1301